MSU RETURNING MATERIALS: P1ace in book drop to LIBRARIES remove this checkout from Ail-(SlI-L. your record. FINES w111 be charged if book is returned after the date stamped below. *1." ‘92"; (E1 m INFLUENCE OF PASTEURELLA HARDLYTICA AND ITS PRODUCTS ON BOVINE NMOPHIL FUNCTION 33' Gary Lee Watson A DISSERTATION Submitted to Michigan State University in partial fulfillnent of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Pathology 1988 5o7603€ ABSTRACT THE INFLUENCE 0F.PASTEURELLA.HAEHOLYTICH.AND ITS PRODUCTS ON BOVINE NEUTROPHIL FUNCTION By GARY'LEE‘UATSON Pneumonic pasteurellosis is a fulminant respiratory disease of ruminants caused by Pasteurella haemolytica (PH) serotype A1. Neutrophil depletion prevents pulmonary injury by Pasteurella haemolytica affirming that pro- ducts produced or released by neutrophils are important factors in the pathogenesis of pneumonic pasteurellosis. To investigate the effects of Pasteurella haemolytica and its products on bovine neutrophil function, neutrophils were isolated from neonatal calves to define and compare: 1) oxidative responses of bovine neutrophils to selected agonists and to Pasteurella haemolytica, 2) effects of Pasteurella haemolytica and Escherichia coli endo- toxins on the oxidative responses of bovine neutrophils, and 3) release of cytoplasmic granules and enzymes by bovine neutrophils in response to agonists and to Pasteurella haemontica. Luminol-enhanced chemiluminescence and superoxide production were measured to quantitate bovine neutrophil oxidative metabolism. Latex particles, phorbol ester, calcium ionophore, and opsonized zymosan were used as agonists. Opsonized zymosan and opsonized live Pasteurella haemolytica enhanced chemiluminescence and superoxide production by neutrophils, but these responses were exceeded by those exposed to latex, phorbol ester and calcium ionophore. Exposure of neutrophils to Pasteurella haemolytica or Escherichia coli endotoxin enhanced superoxide pro- duction in response to phorbol ester, calcium ionophore. and opsonized zymosan, but chemiluminescence was diminished when the agonists were phorbol ester or calcium ionophore. The greatest release of beta-glucuronidase, vitamin BIZ-binding protein and cytosolic lactate dehydrogenase occurred in neutrophils exposed to live-opsonized Pasteurella haemolytica. These results indicate that: 1) live Pasteurella organisms and opsonins facilitate the release of reactive oxygen species and superoxide anion from neutrophils, 2) the production of superoxide, by neutrophils, is enhanced by exposure to trace quantities of endotoxin, and 3) live organisms and opsonins markedly enhance the release of BlZBP (specific granules) partially through the lysis of neutrophils. It is postulated that these events may have major roles in the neutrophil-mediated injury in pneumonic pasteurellosis. To Alga "Shorty" and Marjorie Watson, my parents, and Ellie, my wife ii ACKNOWLEDGEMENTS My deepest and heartfelt love and thanks go to my loving wife Ellie without whose inspiration and support this dissertation could.have never been completed, and to my three beautiful daughters, Dawn, Cherie, and Kris, who have suffered without a father during practice, residen- cies, doctoral research, employment, and lastly thesis writing. Maybe now I can.make up for lost time. Appreciation and special thanks are extended to Dr. Stuart Sleight, my adopted major professor and Dr. Ed Rabinson for their immeasurable efforts and contributions. Thanks go to Dr. Ron Slocombe for the grant supporting this research, and to Dr. Ken.Keahey for supporting my family. Other thanks go to my committee, Drs. Tom Bell, Glenn “axler, and Julie Stickle. Without the technical expertise of Beverly 'Kay' Trosko, and her attention to detail, these 'experimental protocols" could never have been completed. Accolades go to Karen Guarisco who provided hours of effort in experiments, mastering SAS statistics and computers, and for calf care assisted ably by Cathy Berney and the crew of the Pulmonary Laboratory. iii TABLE OF CONTENTS Page List of Tables . . . . . . . . . . . . . . . . . . . . vi List of Figures . . . . . . . . . . . . . . . . . . . viii Introduction . . . . . . . . . . . . . . . . . . . . 1 Literature Review, Part I: Mechanisms of Neutrophil Activation (Transduction). 3 Introduction . . 3 Mechanisms of Neutrophil Transduction. . . . 4 Neutrophil Response Phases . . . . . . . . . . . . 4 Intracellular Second Messengers. 4 Receptor Based Activation. 5 Protein Kinase C . 8 The Respiratory Burst. . . . . 9 Quantitation of Respiratory Burst Activity . . . . 12 Mechanisms of Tissue Damage and Cell Protection. . 13 Mechanisms of Enzyme Release . . . . . . 16 The Role of Endotoxin in Neutrophil Function . . . 20 Literature Review, Part II: The Bovine Neutrophil . . 28 Introduction . . . . . . . . . . . . . . . 28 Bovine Granulocyte Kinetics. . . . . 28 Biochemical Constituents of Bovine Neutrophils: Comparison to Other Species. . . . . . . . . . . . 29 Bovine Neutrophil FUnction Studies . . . . . . . 32 The Role of Opsonins in Bovine Neutrophil Responses . . . . 43 Selenium and Vitamin E and Neutrophil Function . . 45 Endotoxin and Bovine Neutrophils . . . . . . . . . 46 Overview and Conclusions . . . . . . . . . . . . . 47 Chapter 1. Luminol-Enhanced Chemiluminescence and Superoxide Production By Bovine Neutrophils in Response To Selected Agonists: Comparison of the Responses to Pasteurella haemolytica. . . . . . . . . . . . . . . . 52 Introduction . . . . . . . . . . . . . . . . . 52 Materials and Methods. . . . . . . . . . . . . . . 54 Experimental Animals. . . . . . . . . . . . . 54 Preparation of Solutions. . . . . . . . . . . 56 Neutrophil Isolation. . . . . . . . 56 Preparation of Agonists and Reagents. . . . . 57 Chemiluminescence . . . . . . . . . . . . . . 59 Specificity of LECL . . . . . . . . . . . . . 61 iv Superoxide Assay. Effect of the Age of the Calf on Total and Differential WBC Counts . . . . . . . Effect of Blood Handling. . Determination of Agonist Levels . . Effect of Individual Calf Variability and Age On LECL and Superoxide Production . . Comparisons to Pasteurella haemolytica. Results. . Discussion and Conclusions . Chapter 2. The Effects of Endotoxin on the Oxidative Metabolism of Bovine Neutrophils. Comparative Effects of Pasteurella haemolytica and Escherichia coli Endotoxins . . Introduction . Materials and Methods. Chemiluminescence . Superoxide Production . Preparation of Endotoxin. Experimental Design . Statistical Design. Results. . Discussion and Conclusions . Chapter 3. Enzyme Release by Bovine Neutrophils . Introduction . . . . Materials and Methods. Experimental Design. Statistical Analyses . Results. Discussion . Conclusions. Summary. List of References . Page 61 63 63 64 65 70 71 80 95 95 96 97 98 99 100 101 102 109 122 122 124 128 129 129 137 144 146 148 LIST OF TABLES Table Page 1. Mechanisms of Formation of Superoxide Anion . . 14 2. Subcellular Localization of Enzymes and Other Constituents Released by Human Neutrophil . . . 17 3. Stimuli for Granule Exocytosis by Human Neutrophils . . . . . . . . . . . . . . . . . . 19 4. Enzyme Activity in PMNs from Various Species. . 3O 5. Enzyme Activities and Content of Binding Proteins of Granulocytes of Bovine and Human Blood . . . . . . . . . . . . . . . . . . . . . 31 1.1. Selection of Agonist Concentrations for Use in Studies of Luminol-Enhanced Chemiluminescence by Bovine Neutrophils . . . . . . . . . . . . . 66 1.2. Selection of Agonist Concentrations for Use in Studies of Superoxide Production by Bovine Neutrophils . . . . . . . . . . . . . . . . . . 68 1.3. Effect of Weeks in the Study on Total White Blood Cell and Percent Neutrophil Counts. . . . 72 1.4. Luminol-Enhanced Chemiluminescence and Superox- ide Production: Comparison of the Percentage of 02 Responses Inhibited by the Use of Heated Serum for Opsonization of Zymosan . . . . . . . 72 1.5. Bovine Neutrophil Oxidative Responses. Luminol- Enhanced Chemiluminescence: Peak Response, Initial and Peak Response Times . . . . . . . . 74 1.6. Luminol-Enhanced Chemiluminescence: Comparison of Percentage Inhibition of Neutrophil Responses to Agonists by Superoxide Dismutase . . . . . . 74 1.7. Bovine Neutrophil Oxidative Responses: Superox- ide Production in Response to Soluble and Partic- ulate Agonists. . . . . . . . . . . . . . . . . 76 vi 1.8. 2.1. Effect of Calf and Weeks in the Studies (Age) on Luminol-Enhanced Chemiluminescence and Superoxide Production Responses to Soluble and Particulate Agonists. . . . . . . . . . . . . . . . . . . . 76 Comparative Endotoxin Studies: Effects of Pasteurella haemolytica and Escherichia coli Endotoxins on the Oxidative Metabolism of Bovine Neutrophils. . . . . . . . . . . . . . . 103 vii LIST OF FIGURES Figure Page 1.1. 1.2. 1.3. 1.4. 1.5. 1.6. Overview of the Mechanisms of Receptor-Mediated Activation of Neutrophils: Diagrammatic Repre- sentation of the Early Events of Neutrophil Acti- vation By Chemotactic Factors and the Postulated Points at Which Pertussis Toxin Acts. Graphic Representation of Experiments Performed In the Selection of Treatments and Agonist Concen- trations for Use in Studies of Luminol-Enhanced Chemiluminescence (Table 1.1) by Bovine Neutrophils. . . . . . . . . . . . . . . . . . . . . . . . . 67 Graphic Representation of Experiments Performed in the Selection of Treatments and Agonist Concentra- tions for use in Studies of Superoxide Production (Table 1.2) by Bovine Neutrophils. . . . . . . . . . . . . . . . . . 69 Bovine Neutrophil Luminol-enhanced Chemilumi- nescence (LECL): Comparison of the Peak Responses, Initial Response Times, and Peak Response Times Induced by Soluble and Particulate Agonists. . . . . . . . . . 75 Bovine Neutrophil LECL: Comparisons of the Peak Responses Induced by Soluble and Particulate Ago- nists and Those Induced by Pasteurella haemoly- tica, Formyl-Methionyl-Leucyl-Phenylalanine (FMLP) and Zymosan Activated Serum (ZAS). . . . . . . . . . . . . . . . . . 78 Bovine Neutrophil LECL: Comparisons of the Slope Responses Induced by Soluble and Particulate Agonists and Those Induced by Pasteurella haemolytica, FMLP, and ZAS. . . . . . . . 79 Bovine Neutrophil Superoxide Production: Comparison of Responses Induced by Soluble and Particulate Ago- nists and Those Induced by Pasteurella haemoly- tics, FMLP, and ZAS. . . . . . . . . . . 81 viii Comparison of the LECL Peak Responses of Bovine Neutrophils Exposed to Trace Quantities of Pasteurella haemolytica Endotoxin (PHzLPS) and Escherichia coli Endotoxin (EC:LPS) upon Stim- ulation by Soluble and Particulate Agonists. . . . . . . . . . . . . . . . . . . . 105 Comparison of the LECL Slope Responses of Bovine Neutrophils Exposed to Trace Quantities of PH:LPS and EC:LPS Upon Stimulation by Soluble and Partic- ulate Agonists. 106 Comparison of the LECL Initial and Peak Response Time Responses of Bovine Neutrophils Exposed to Trace Quantities of PHzLPS and EC:LPS Upon Stimulation by Soluble and Particulate Agonists. . . . . . . . . . . . . . . . . . . . . . . . . 108 Comparison of the Superoxide Production of Bovine Neutrophils Exposed to Trace Quantities of PH:LPS and EC:LPS Upon Stimulation by Soluble and Particu- late Agonists. 110 Enzyme Release 1: Comparison of the B-glucuronidase (BG) and Lactate Dehydrogenase Release (LDH) of Bovine Neutrophils Upon Stimulation by Soluble and Particulate Agonists. 130 Enzyme Release 2: Comparison of the BG and LDH Release of Bovine Neutrophils Upon Stimulation by Pasteurella haemolytica. 132 Enzyme Release 3: Comparison of the BC and LDH Release by Bovine Neutrophils in Response to Stimu- lation by Soluble and Particulate Agonists and by Pasteurella haemolytica. 134 Enzyme Release 4: Comparison of the BG, LDH, and Vitamin B12-Binding Protein (B12BP) Release by Bovine Neutrophils in Response to Stimulation by Soluble and Particulate Agonists and by Pasteurella haemolytica. Part I: BC Release. 135 Enzyme Release 4: Comparison of the BG, LDH, and B12BP Release by Bovine Neutrophils in Response to Stimulation by Soluble and Particulate Agonists and by Pasteurella haemolytica. Part II. LDH Release. 136 ix 3.6. Enzyme Release 4: Comparison of the BG, LDH, and B12BP Release by Bovine Neutrophils in Response to Stimulation by Soluble and Particulate Agonists and by Pasteurelia haemolytica. Part III. B12BP Release. 138 INTRODUCTION Pneumonic pasteurellosis caused by the gram negative bacterium Pasteurella haemolytica, has a primary role in the bovine respiratory disease complex. Slocombe et. a1.129 recently determined that this disease is neutro- phil-mediated. Neutrophil-mediated injury has been attri- buted to the production of reactive oxygen species, includ- ing superoxide anion and its metabolites, the release of enzymes from cytoplasmic stores, and the production of metabolites of arachidonic acid (prostaglandins and leuko- trienes). Due to the complexity and interrelationships of diverse pathways active in vivo, it was felt necessary to investigate the interactions of Pasteurella and neutrophils by in vitro methods. In my experiments I decided to investigate the produc- tion of oxidative metabolites and enzyme release, by neutrophils, in response to the Pasteurella organism. Studies of the oxidative metabolism of bovine neutrophils were designed to investigate the capacity of stimulated bovine neutrophils to produce superoxide anion and other reactive oxygen metabolites. To characterize the magnitude of bovine neutrophil responses it was necessary to define a battery of agonists which would provide consistent and maximal responses in bovine neutrophils. It was hoped that this series of agonists would also provide a ”frame of reference" for planned comparisons of the responses to the Pasteurella organism. A recent publication by Guthrie et. al.77 described a phenomenon induced by the exposure of neutrophils to trace quantities of bacterial lipopolysaccharide (endotoxin), termed "priming", where exposed cells markedly increased the production of superoxide in response to a diverse series of agonists. Since the Pasteurella organism has the capability of producing endotoxin, a series of experi- ments were performed to investigate whether Pasteurella endotoxin might induce similar responses in bovine neutro— phils. These experiments utilized the agonists defined in the oxidative metabolism protocols as stimuli. Lastly, the same agonists used in the investigations of oxidative metabolism and the effects of endotoxin were used to determine the magnitude of enzyme release by bovine neut- rophils. Subsequently, the responses to these agonists were compared to those initiated by exposure of neutrophils to the Pasteurella organism. It was hoped that these in vitro experiments would provide information about the role of neutrophil products in the pathogenesis of pneumonic pasteurellosis, and serve as a basis for future investigations into means and methods for modulating or preventing the disease. LITERATURE REVIEW: PART I: MECHANISMS OF NEUTROPHIL.ACTIVATION (TRANSDUCTION) INTRODUCTION: Within the past five years more than 25,000 publications have emphasized various aspects of neu- trophil function. The majority of these publications cen- tered upon the cellular biology of the neutrophil and at- tempted to define the mechanisms of neutrophil-mediated in- jury. Few of the studies were related to the general fea- tures of the bovine neutrophil or to the specifics of neutrophil-mediated diseases in the bovine species. This literature review will be divided into two portions: a sum- mary of the currently understood mechanisms of neutrophil transduction or activation and a review of the current literature which specifically applies to the bovine neutro- phil. A major assumption is that while the bovine neutro- phil may possess phenotypic differences in enzymic compo- nents, and perhaps in biological behavior, the biochemical nature of the transduction mechanisms, although not in actuality defined, are essentially similar. While the prin- cipal thrust is in the exploration of (bovine) neutrophil oxidative metabolism and enzyme release, due to the com- plexity of interrelated pathways, a review of those 4 aspects of neutrophil function cannot be separated dis- tinctly from other functions initiated simultaneously. MECHANISMS OF NEUTROPHIL TRANSDUCTION: There are 1-4 many excellent reviews of neutrophil function and the 5'6, all providing role of the neutrophil in inflammation a basis from which to initiate a review of the recently de- fined aspects of neutrophil activation or transduction. NEUTROPHIL RESPONSE PHASES: The neutrophil responses to external stimuli occur in three phases.6 Phase I occurs through ligand binding resulting in altered membrane topography with exposure of surface sialic acid residues, glycoproteins and receptors. Phase II involves the opening of ion channels or molecular pores with exposure of ion pumps, transport processes, or surface enzymes. Altered ionic permeability and cation fluxes may trigger phase III directly or indirectly through the formation of second messengers. The phase III response leads to activation of cryptic enzyme systems which control or mediate membrane fusion. Phase II ligands preferentially activate the energy producing pathways including glucose transport, glycolysis and glycogenolysis whereas phase III ligands activate energywasting or catabolic pathways. INTRACELLULAR SECOND MESSENGERS: Central to the acti- vation of neutrophils are the intracellular second messen- 2+ gers, the divalent cation Ca and products of polyphos- phoinositide metabolism.l"7'9 The complexity of this 5 activation is exemplified by the fact that phospholipase C, the enzyme necessary for formation of the two primary se- cond messengers diacylglycerol (DAG) and inositol triphos- phate (1P3), is itself a Ca2+-dependent enzyme. DAG in combination with Ca2+ and a phosphatidylserine (PS) sub- strate are the endogenous activators of the intracytoplas- mic enzyme protein kinase C (PKC).7 1P3 is important in the release of intracellular calcium stores from the endoplasmic reticulum.9 Recently studies have defined the presence of a 3-kinase, activated by a calmodulin- /Ca2+-comp1ex, which converts 1P3 to inositol tetrakis- phosphate (IP4) which appears to modulate the entry of calcium from the outside of the cell, through effects on calcium pumps. Also identified was a 5-phosphomonoester- ase, activated by protein kinase C, which catalyzes the for- mation of 1P2, thereby terminating the second messenger response.lo Divalent calcium has a major role as a co- factor for intracytoplasmic enzymes including phopholipase C and protein kinase C, as mentioned, but also for other enzymes important in stimulus-mediated events, including phospholipase A2, calmodulin, and Ca2+ -requiring thiol proteinases (calpains). Calcium is also required for micro- tubule assembly and for actin and myosin interaction, neces- sary for leukocyte motility. This triumvirate also has a role in the migration of lysosomal granules to form phagoly- sosomes and for the migration of secretory granules to the cell surface.6 6 RECEPTOR BASED ACTIVATION: The response of neutro- phils to Chemotactic factors [formyloligopeptides (FMLP)], anaphylatoxins (C3b, C5a), leukotrienes, and platelet- activating-factor (PAF), are mediated through GTP-binding regulatory proteins, effected through the activation of phospholipase C (PLC).11 A current perspective of recep- 1.11 An intra- tor-mediated events is included as figure cellular pool of cell surface receptors is currently felt to reside within the membranes of specific gran- uleslz'ls , where the process of degranulation results in the addition of new plasma membrane and is associated with the mobilization of surface receptors.14 Receptor stimu- lation leads to the hydrolysis of membrane polyphosphoinosi- tides into DAC and IP3. Pertussis toxin (PT) has been used as a probe of guanyl regulatory proteins since it ribo- sylates and blocks the inhibitory arm (N1) of the guanyl proteins. This blockage acts at steps responsible for re- ceptor-induced calcium mobilization and for the activation of protein kinase 0.11 However, PT does not affect the increase of free intracellular Ca2+ induced by platelet activating factor (PAF), but completely inhibits enzyme se- cretion.16 PT almost completely abolishes the breakdown of membrane phosphoinositides, suggesting that Ni is in- timately coupled to the activation of PLC. PT also pre- vents the release of arachidonic acid caused by Chemotactic factors, which may be due to: a) lack of activation of 6A Figure 1. Overview of the Mechanisms of Receptor- Mediated Activation of Neutrophils: Diagrammatic Representation of the Early Events of Neutrophil Activation by Chemotactic Factors and the Postulated Points at Which Pertussis Toxin Acts. PLA2 and PLD are abbreviations for phospholi- pases A and D respectively. PtIns, PtIns4(P), and Ins(4,5)P3 represent phosphatidylinositol and its mono- and trisphosphate forms respective ly. From ref 15. Eczcdiagufl . E ----------------1 o .3 uuuuuuuuuuuuuuuu um 300,000, as determined by 40 ultrafiltration. Czuprynski et. a1.131 determined that ingestion of virulent Pasteurella by neutrophils was serum dependent and mediated principally by heat stable opsonins (antibodies). Killing occurred within 1-4 hours of ingestion, with reduced neutrophil viability at bacteria to cell ratios of 100:1. The ingestion of opsonized PH stimulated a marked CL response. O'Brien and Duffus132 compared the susceptibilities of bovine leukocytes to the PH cytotoxin and determined that neutrophils were the most susceptible, macrophages less so, and that bronchoalveolar macrophages from adult animals were the most resistant. They also determined that sub-cytolytic concentrations of cytotoxin did not impair killing of virus infected cells by macrophages. Chang et. al.133 used luminol-dependent chemilumines- cence (LDCL) to study the responses of neutrophil prepara- tions from 4 species of ruminants and 6 species of nonrumi- nants to opsonized and nonopsonized live and dead PH and Staphylococcus aureus. Opsonized organisms enhanced the reponse in all species. Living PH and culture super- natants inhibited the neutrophil LDCL for all ruminants, 134 the but not for nonruminants. In a companion study LDCL responses of bovine neutrophils to living and killed PH, Pasteurella multocida, and Escherichia coli were evaluated. Live and opsonized bacteria enhanced re- sponses, but with PH the LDCL precipitously declined after 41 a peak response at 10 minutes. Similar responses were seen with culture supernatants. The authors suggested that the LDCL-inhibition might be used to quantitate the concentra- tions of cytotoxins produced by PH. A subsequent re- port135 compared the utilization of the LDCL-inhibition (LDCLI) assay to trypan blue dye exclusion (TBDE) and to 51Cr-release assays (CRA) and determined the relative sen- sitivities of the assays in detecting biological activity of PH leukotoxin. These studies revealed the LDCLI assay to be 17 times more sensitive than the CRA.and 2,480 times more sensitive than the TBDE. The authors proposed the uti- lization of chemiluminescence inhibitory units (CIUs), which were defined as the quantities of cytotoxin required to abbrogate the chemiluminescence response of 106 neutro- phils within 20 minutes of incubation. Serial dilutions of cytotoxin were used to determine the least amount that would inhibit cell response. Other investigators have also used the production of chemiluminescence in the study of other parameters of bo- vine neutrophil function. Forsell136 used the generation of CL and phagocytosis assays to determine that the wood preservative pentachlorophenol had no effect on neutrophil function. Phillips et. al.137 investigated the effects of dexamethasone administration on neutrophil CL and found that there were no significant effects on phagocytes in whole blood, but that single or multiple doses of the drug 42 enhanced the CL response of purified cells. A companion report138 determined that PGEl, PGEZ, indomethacin, and histamine suppressed the CL response and that the suppres- sion was dependent upon the continuous presence of the com- pounds. However, the inhibition of CL induced by indometha- cin persisted even after removal of the compound from the culture media. Few studies of bovine neutrophil oxidative metabolism have quantitated either hydrogen peroxide or superoxide anion production or measured oxygen consumption, all con- sidered to be measures of respiratory burst activity. 139 compared the production of Korhonen and Reiter H202 by blood-derived and milk-derived neutrophils. Hydrogen peroxide could not be detected in the presence of bovine serum, fetal calf serum, or milk whey unless sodium azide (NaNz) was added to antagonize endogenous catalase and lactoperoxidase. Bovine serum or milk whey was requir- ed for phagocytosis of live organisms and for hydrogen per- oxide production. IgGZ and to a lesser extent IgGl, but not SIgA and IgM, stimulated hydrogen peroxide release inde- pendently of phagocytosis. Neutrophils isolated from peri- pheral blood were more active in hydrogen peroxide produc- tion than were those isolated from milk. In 1986, YOung and Beswick140 compared the responses of bovine, porcine, ovine and human neutrophils to serum-treated zymosan and found significantly less superoxide production and oxygen 43 consumption of animal species compared to human cells. The differences in responses did not appear to be influenced by the origin of the serum. Human neutrophils responded to FMLP, but the neutrophils of the animal species did not. THE ROLE OF OPSONINS IN BOVINE NEUTROPHIL.RESPONSES: The quantities of opsonins in serum are of major importance in the facilitation of phagocytosis and involve the levels 141'144 as well as products of specific immunoglobulins, of the complement cascade pathways.145'148 Complement products such as C3b, derived from the classical pathway, and 65a, derived from alternate and classical pathways, are important not only in their roles as chemotaxins, but when deposited on a surface or particulate entity they enhance neutrophil ingestion through the interaction with cell sur- face receptors. While the pathways of complement activa- tion and their role in the process of Opsonization are well understood, the participation of bovine immunoglobulin sub- classes in this process remains a subject of confusion and debate. Principal roles for IgGl, IgGZ, and IgM in the fix- ing of complement and the enhancement of phagocytosis have been described by several investigators.1£‘1'143 In neona- tal calves, early protection against infection is derived from the absorption of colostral immunoglobulins with IgGl being preferentially absorbed over IgGZ by an approximate ratio of 7:1.141 While both IgGl and IgGZ can fix comple- ment by means of the classical pathway, it was determined 44 that only IgGZ fixes it by means of the alternate path- 141'143 and facilitates phagocytosis.143 Since the way defense against gram negative organisms, such as Pasteur- ella, is principally dependent upon the alternate path- way, selective deficiencies of either IgGZ or alternate pathway components might render neonatal calves more suscep- tible to bacterial disease. While immunoglobulins are pas- sively acquired, complement factors are not, and normal (adult) levels are not reached until approximately six months of age.1“5'148 Lombardo et. al.149 compared the effects of colostral ingestion on leukocyte counts, and found that calves fed colostrum had higher total leukocyte counts and higher per- centages of neutrophils than did colostrum-deprived calves. For functional studies of bacterial ingestion by neutro- phils, E. coli organisms were opsonized with three serums: l) neonatal pre-colostral, 2) autologous post-colos- tral and 3) adult, with monitoring of neutrophil responses by quantifying oxygen uptake. The highest oxygen uptake rate was obtained by the use of adult serum, with autolo- gous intermediate, and neonatal the lowest. The use of neo- natal serum for opsonization produced no significant differ- ences in the responses of neutrophils derived from colos- trum-fed (CF) and colostrum-deprived (CD) calves. However, the use of autologous or adult sera significantly increased oxygen uptake. Activity of the HMP shunt and glycolysis 45 were examined by determining the oxidation of labeled 14C-glucose to COZ and lactate. There were no signifi- cant differences between CF and CD in HMP activity or in lactate formation. The effects of maternal protein-energy malnutrition and cold stress on neutrophil function and sera were studied by Woodard et. a1.150 Nutritional deficiencies in the dam had little effect on in vitro bactericidal activity of neutrophils or pre-colostral sera taken at birth. Neutro- phils obtained at birth killed Staphylococcus aureus but not Escherichia coli, when incubated with heated or unheated autologous serum. With pre-colostral serum, neutrophils from 3-day-old calves were no more active in destroying bacteria than those of newborn calves. The addi- tion of day 3 serum enabled day 3 neutrophils to destroy E. coli, but this was not influenced by heat treatment of sera. Maternal protein deficiency increased the destruc- tion of E. coli by day 3 neutrophils and sera. There were no differences in the bactericidal effects of neutro- phils or sera from cold stressed calves, nor were cold stress-nutritional stress interactions detected. SELENIUH AND VITAMIN E AND NEUTROPHIL FUNCTION: The necessity for both of these compounds as major components of intracytoplasmic and membranous defense against peroxida- tive attack from oxygen and lipid derived radical species, has been addressed in Part I of this review. At high 46 levels of supplementation vitamin E enhances both humoral and cellular immunity by means of an, as yet undefined, en- hancement of the phagocytosis of antigens.66 Adequate se- lenium also appears to be necessary for humoral re- sponse.151 Selenium deficienciesl'52’153'156 while not affecting the ability of neutrophils to phagocytize yeasts (Candida albicans)155 or bacteria (Staphylococcus aureus)152’157, markedly diminish the ability of cells to kill ingested organisms.157 In neutrophils, the pro- duction of superoxide and hydrogen peroxide is initiated by activation of the respiratory burst enzyme NADPH-oxidase in which the rate limiting step is the availability of reduc- ing equivalents (NADP) supplied by the activation of the HMP shunt. Sources of NADP include the superoxide forming reaction and catabolism of hydrogen peroxide by means of the glutathione reductase and glutathione peroxidase sys- tem. Since selenium is required for activity of the selen- ium-dependent form of glutathione peroxidase, and since this enzyme represents the primary means of disposal of hy- drogen peroxide in bovine neutrophilsgg, impaired func- tion limits the production of NADP, reducing the generation of superoxide, and thereby altering killing ability.152 ENDOTOXIN AND BOVINE NEUTROPHILS: The principal in vivo studies in the bovine species have used the intramam- mary administration of endotoxin to determine the effects of and similarities to naturally occurring endotoxic 47 mastitis. Endotoxin administration induced significant in- creases in milk whey IgGl, IgG2, IgM, and IgA with the greatest relative increases seen for IgGZ. Cows with endo- toxin-infused quarters, treated systemically with flunixin meglumine, had significantly decreased milk whey concentra- tions of IgGl and IgM. While endotoxin treated quarters seemed to enhance the phagocytosis of Staphylococci by milk PMNs, there were no differences in the flunixin or saline treated (control) animals.158 In mastitis, neu- trophil influx is immunoglobulin related, but few studies have attempted to determine whether endotoxin has direct effects on the bovine neutrophil. One study determined that a principal effect of LPS was the inhibition of cellular migration. In the only report dealing with the direct, in vitro effects of LPS on neutrophils, Confer and Simons159 used Pasteurella haemolytica LPS and determined that PH LPS itself was not toxic to neutrophils, and had little effect on neutrophil random migration. Phagocytosis of labeled Staphylococcus aureus was decreased by low (2.5 113/106 cells) and high (65 ug/106 cells) concentrations of LPS, whereas moderate levels (5-25 ug/106 cells) increased the rate of ingestion. PH LPS also enhanced the NBT reduction of exposed cells. OVERVIEW AND CONCLUSIONS: Part I of this review con- sisted of an attempt to summarize, in a concise manner, the current thought and perspectives of neutrophil cellular 48 biology and mechanisms of transduction. While the aspects that pertain to the major postulated mechanisms of neutro- phil-mediated injury: 1) products of oxidative metabolism, 2) enzyme release, and 3) release of arachidonate metabo- lites have changed little over the years, the increased sophisticatiOn of research efforts, which has enabled inves- tigators to dissect cellular responses, has led to an in- creased level of understanding of the complexity of the bio- chemical events leading to neutrophil-mediated tissue in- jury. This often overwhelming mountain of information on the cellular biology of neutrophils have been a result of the desire by all investigators to determine the solitary mechanism for each response, the mechanistic approach to in- vestigation. Neutrophils, however, are a biological match for the appetites of these micromolecular investigators, in that no solitary stimulus induces a solitary response. In contrast to the mechanistic methods of part I, studies of the biology of the bovine neutrophil are functionalistic. Rather than attempting to define the mechanism of a re- sponse, they have sought instead to define the response. In comparison to the techniques used in Part 1, many of our techniques are comparatively primitive in nature. Only time will tell whether the mechanistic approach or function- alistic approach will provide the ultimate benefits to the species we have chosen to study. 49 In subsequent chapters, I will describe the results of my research into the role of neutrophil products in pneumon- ic pasteurellosis of ruminants. These experiments were de- signed to determine: 1) whether products of bovine neutro- phil oxidative metabolism might have a role in the patho- genesis of the disease, 2) whether bacterial lipopolysac- charide (endotoxin) had effects on bovine neutrophil oxida- tive metabolism that might contribute to the pathogenesis of the disease, and 3) whether the release of cytoplasmic granules and enzymes by stimulated neutrophils might have a role in the pathogenesis of the disease. In Chapter 1, I will describe the means and methods used to define the oxidative metabolism of bovine neutro- phils. For these experiments I chose to use chemilumines- cence, enhanced by luminol, (LECL) and superoxide produc- tion to quantitate the neutrophil responses. To provide a basis for future comparisons of bovine neutrophil oxidative responses, I chose to use a series of agonists which induce neutrophil oxidative metabolism by different mechanisms. To ensure that the concentrations or quantities of these agonists maximally activated bovine neutrophils a series of dose response studies was intitially performed. The re- sults of these experiments led to the definition of a bat- tery of agonists which were subsequently used in compari- sons of the oxidative responses of neutrophils to the Pasteurella organism. The spectrum and concentrations 50 of "defined agonists" were used in later experiments to determine the effects of endotoxin on oxidative responses of bovine neutrophils and in studies defining enzyme re- lease. In Chapter 2, interest into the possible effects of endotoxin on neutrophil oxidative metabolism was fueled by a recent report of increased superoxide production after exposure of neutrophils to trace quantities of endo- toxin.77 I questioned whether endotoxin might have simi- lar effects on bovine neutrophils. To answer this question I used trace quantities of two different endotoxins, E. coli 055:B5 endotoxin and P. haemolytica endotoxin. I sought to resolve the premise that the biological effects of endotoxins differ little between bacterial species of origin and to determine whether the effects of Pasteur- ella endotoxin might induce a response that was specific for bovine neutrophils. Through the use of the previously defined agonists, I hoped to determine whether endotoxin- induced effects were specific for some mechanisms of induc- tion but not for others. In Chapter 3, I used the release of B-glucuronidase, a marker of primary granule release, vitamin-Blz-binding protein, a marker of secondary granule release, and lactate dehydrogenase, a measure of neutrophil death/lysis, to de- termine the spectra of enzyme release in response to the previously defined agonists. The results of these studies 51 provided a basis for comparing the amounts of these markers released by exposure of neutrophils to P. haemolytica. I hoped to determine whether the selective release of gran- ules subtypes, induced by the organism, might have a signi- ficant role in pneumonic pasteurellosis. In summary, I hoped to interpolate my understanding of the mechanisms of neutrophil transduction, from experiments suggested by the reports of other investigators, into a series of experiments which would allow me to define the mechanisms of neutrophil-mediated injury induced by the Pasteurella organism. In definition of these mechan- isms, I hoped to provide a basis for future studies into the modulation of the disease. CHAPTER 1. LUHINOL ENHANCED GHEHILUHINESCENCE AND SUPEROXIDE PRODUCTION BY BOVINE NEUTROPHILS IN RESPONSE TO SELECTED.AGONISTS: GOHPARISON OF THE RESPONSES T0.PASTEURELLA.HAEHOLYTIGA Ingrgggggign: Neutrophils respond to soluble and particu- late agonists by the sequential processes of chemotaxis, phagocytosis, degranulation and increased metabolic ("re- spiratory burst") activity}.3 As a consequence of neu- trophil activation, pulmonary injury may occur. Pulmonary injury by neutrophils involves three major mechanisms: the generation of oxygen-derived free radicals, the synthesis and release of proteolytic enzymes from specialized cyto- plasmic granules, and liberation of arachidonic acid-de- rived metabolites.1'3’5’160'164 No unifying theory ade- quately explains differences in the mechanisms of neutro- phil-mediated injury, although a basic understanding of these processes is known. In some diseases tissue injury is caused by secretion of proteolytic enzymesl65'168, yet in others injury may be lessened by blockers of oxygen radi- cal formation or by inhibitors of arachidonic acid metabol- ism.169'176 52 53 The release of oxygen-derived radicals, enzymes and arachidonate metabolites may occur by means of diffusion, secretion, cell lysis, reverse endocytosis ("frustrated phagocytosis”), or by regurgitation during feeding when phagosomal contents escape prior to closing of phagocytic vacuoles.1 Adverse effects of neutrophils have been demonstrated in acute pulmonary diseases such as pulmonary thromboembolism, endotoxic shock, adult respiratory dis- tress syndrome (ARDS) and immune complex interstitial pneu- monias.177"181 Neutrophil depletion in sheep182 and cattle129 has been used to define the role of neutrophils in the mediation of lung injury in endotoxemia and in bac- terial pneumonia. Although neutrophils have been incrim- inated in the initiation of pulmonary injury in pneumonic pasteurellosis129 , there is no information regarding which neutrophil products are damaging in this disease syn- drome. Experiments in our laboratory in which calves were pretreated with polyethylene glycol (PEG)-bound catalase to catabolize the release of hydrogen peroxide, or with cyclo- oxygenase inhibitors did not diminish the severity of ex- perimentally-induced Pasteurella pneumonia (unpublished data). Studies by Gennaro et. 31.99 determined that stimulat- ed bovine neutrophils had a greater capacity to produce superoxide (02-) and hydrogen peroxide (H202) than did human neutrophils responding to the same stimuli. To 54 further evaluate these responses, two measures of neutro- phil oxidative metabolism were chosen: luminol~enhanced chemiluminescence (LECL) and superoxide production. Concen- trations of agonists which induced maximal neutrophil oxida- 53’54 were used as a tive metabolism in other species guide for dose response studies and subsequently in experi- ments to determine the magnitude of bovine neutrophil LECL and superoxide production. The spectrum of agonists se- lected to evaluate the responses of bovine neutrophils to differing mechanisms of transduction (activation) included: 1) non-opsonized particulates (latex); 2) opsonized particu- lates [opsonized zymosan (02)]; 3) an exogenous activator of protein kinase C [phorbol myristate acetate (PMA)]; and 4) an inducer of increased intracellular calcium [calcium ionophore, A23187 (CI)]. The increased susceptibility of neonates to disease has been documented and partially ascribed to defective neutro- phil function in human infant8183'184 and calves.127’149’152 A series of experiments was performed to evaluate whether the neutrophils of calves of different ages had differences in LECL or superoxide production re- sponses which might be attributed to induction by specific agonists. In summary, this series of experiments was designed to: 1) define some of the factors influencing bovine neutrophil oxidative responses, 2) define the concentrations of SS agonists which maximally stimulate bovine neutrophil oxida- tive metabolism, 3) quantitate the magnitude of LECL and 02- production of bovine neutrophils stimulated by de- fined agonist concentrations, 4) determine the effects of age on the oxidative metabolism of bovine neutrophils stim- ulated by defined concentrations of agonists, and 5) com- pare the oxidative responses stimulated by agonists with those stimulated by Pasteurella haemolytica. MATERIALS AND METHODS: e en : Neonatal Holstein bull calves, less than 1 week of age, were obtained from a large local dairy farm and had received colostrum. On arrival at MSU the calves were weighed, given an oral rotavirus and corona- virus vaccine“, and injections of vitamin E/seleniumb and vitamins A, D, and EC. Blood samples were collected in EDTA tubesd for total and differential white blood cell (WBC) counts. Twice daily, body temperature, pulse and respiratory rates (TPR), appetite, attitude, and charac- ter of feces were recorded. The calves were housed in a stall in a research ward of the Veterinary Clinical Center (VCC), fed milk replacere twice daily and exercised once daily. The calves were acclimated for one week before use aCalf Guard, Norden Laboratories, Lincoln, NE. bBo- Se, Schering Corporation, Kenilworth, NJ. cVitamin ADE Injection, W. A. Butler Co. , Columbus, OH. acutainer Tubes, Becton- Dickinson, Inc, Rutherford, NJ. eFresh Start, Vita Plus Coporation, Madison, WI. 56 and during this time several aliquots of serum were har- vested for later use in the Opsonization of particulate agonists. Each calf was used for three weeks. Preparatign_gf_§glutign§: Hank's balanced salt solu- tion (HBSS) and phosphate buffered saline (PBS) with and without calcium and magnesium were prepared from concen- trated stock solutions‘53’54 Triple deionized water (TDW) was phosphate buffered (0.0132M)102. All solutions were adjusted to a pH of 7.2 and stored at 4 C. Acetate- citrate-dextrose (ACD) anticoagulant was prepared by the large animal pharmacy of the VCC and kept at 4 C. All solu- tions were sterilized by filtration prior to storage and prepared as needed. e o o : Venous blood was collected in syringes containing ACD (lml per 6.6 ml of blood), trans- ferred into blood tubes, and placed in crushed ice. (see "Effect of blood handling", Materials and Methods) Total WBCs were counted with an automated cell counterf, blood smears were stained with a modified Wright-Giemsa stains, and manual differential WBC counts were performed prior to neutrophil isolation. Total and differential (% neutro- phil) counts were recorded. Procedures for isolation of neutrophils were modifications of previously described tech- niques.102'106 Briefly, the blood was centrifuged at ICoulter Counter ZBF, Coulter Electronics, Hialeah, FL. gCamco Quick Stain, Cambridges Chemical Products, Inc., Detroit, MI. 57 1000 X g for 20 minutes in 16 X 125 mm glass tubes, and the plasma, buffy coat and l/4 to 1/3 of the red blood cell layers were aspirated. Erythrocytes were lysed by the addi- tion of TDW (two times the cell volume). The tubes were rapidly but gently inverted 20 times, and an equal volume of 2X PBS (without Ca2+ and Mg2+) was added to restore isotonicity. The cells were centrifuged at 250 X g for 10 minutes, the supernatant was aspirated, and the cells were resuspended in calcium and magnesium-free HBSS. A second lysis, centrifugation and resuspension were then performed. Total and differential WBC counts and trypan blue dye exclu- sion tests for viability were performed on the isolated cells. Test tubes containing the isolated cells were stored in crushed ice until used. All assays were per- formed within six hours of blood collection to ensure maxi- mal cell viability. re a at o 0 ts ea e t : Luminolh (S-amino-2,3-dihydro- 1,4-phthalazinedione) was prepared fresh weekly in PBS with Ca2+ and Mg2+ at a stock con- centration of 10'3 M, adjusted to pH 7.2, and allowed to dark adapt at 4 C for a minimum of 72 hours. Phorbol l2-myristate l3-acetateh’i (PMA) was dissolved in dime- thylsulfoxide (DMSO) at a concentration of 2 mg per ml, and aliquots were stored at -70 C. Calcium ionophoreh (CI) nSigma Chemical Company, St. Louis, MO. 1CMC Cancer Research Chemicals, Brewster, NY. 58 was dissolved in DMSO to obtain a stock solution of 2 X 10'2 M and stored at -20 C. Formyl-methionyl-leu-cyl- -phenyl-alanine (FMLP)h was dissolved in DMSO at a con- centration of 10"2 M, with aliquots stored at -70 C. Zymosanh (10 mg/ml) was suspended in HBSS (with Ca2+ and Mg2+), and opsonized by adding an equal amount of serum to the samples which were placed in a shaking water bathJ at 37 C for 30 minutes. The 02 was washed twice in HBSS and centrifuged at 4 C, with aliquots stored at -20 C. Zymosan-activated serum (ZAS) was prepared by the addi- tion of equal volumes of serum to a zymosan suspension (10 mg/ml) and incubated for 45 minutes in a shaking water bath at 37 C. Post-incubation, the tubes were centrifuged at 1000 X g for 15 minutes at 4 C, and the serum was decanted and heat inactivated at 56 C for 30 minutes. Aliquots were stored at -70 C. Latex particlesh, 0.81 micrometer, were mixed thoroughly in a vortex mixer prior to use. Cyto- chrome C (Horse heart, type VI)h was dissolved in HBSS at 7.5 mg per m1 (0.6mM) for immediate use. Superoxide dismu- tase (SOD)h’k was dissolved in HBSS at 1000 units of activity per ml. The enzyme solution was stored at -20 C, and thawed just prior to use. Stock cultures of Pasteurella haemolytica serotype Al were inoculated into BHI broth or onto blood agar plates JDubnoff Metabolic Shaking Incubator, American Scientific Products, McGaw Park, IL. DDI Pharmaceuticals, Mountain View, CA. 59 and grown at 37 C for 48 hours. The bacteria in broth tubes were centrifuged (1000 X g), the broth was decanted and discarded, and the bacteria were resuspended in HBSS with Ca2+and Mg2+ by means of a vortex mixer. The wash- ing process was repeated twice. Bacterial colonies from blood agar plates were removed with a sterile plunger from a tuberculin syringe and by washing with HBSS. The bacter- ial suspensions were centrifuged at 1000 x g. After wash- ing, the bacterial organisms were resuspended in HBSS and adjusted to an optical density of 0.800 at 541 nm in a double beam spectrophotometer.1 The bacterial suspen- sions were divided into two tubes, with one maintained at 37 C and the other killed by incubation at 56 C for one hour. A portion of the live and dead bacteria were opson- ized by the addition of equal volumes of autologous serum and incubated at 37 C for 30 minutes in a shaking water bath. Live and dead organisms (unopsonized) were included to ensure that all organisms received the same physical .treatments. All suspensions were washed twice and centri- fuged (1000 x g) at 4 C for 10 minutes before final resus- pension.r The final preparations were maintained at room temperature. 1 ce : Chemiluminescence (LECL) was mea- sured in a liquid scintillation counterm calibrated with lShimadzu UV-260, Shimadzu Instruments LTD, Tokyo, Japan. mBeckman LS7500, Beckman Instrument Co., Fullerton, CA. 60 a tritium standard just prior to use. For each agonist, luminol and PMN's (l X 106) were added to a polyethylene minivialn, mixed by inversion, placed into a 20 ml poly- ethylene counting vial and lowered into the counting cham- ber. After three baseline counts, the agonist was added, and the vial was inverted five times and returned to the counting chamber for 15 minutes. The end volume in all vials was 1.0 ml. Each reaction vial was counted at 0.1 minute intervals for 15 minutes. To ensure that the time after isolation of cells did not influence experimental re- sults, the sequence of agonists was changed daily. Adjust- ment of the instrument's sigma error from 2% to 0.01% made it possible to perform all assays utilizing a 70 uM concen- tration of luminol, which enhanced the detection sensitiv- ity of emitted LECL without exceeding the instrument's counting linearity. All counts were obtained with the in- strument in an in-coincidence mode. The instrument counted [in counts per minute (CPM)] and printed five data points per minute (of elapsed time). For analysis of data, the five data points were averag- ed and log-transformed. Peak LECL, initial response time, and time to peak response were recorded. Peak LECL was de- fined as the highest CPM recorded during the 15 minute as- say time period. Initial response time (IRT) was defined as the time (in minutes) when the average of 5 counts more nBeckman Instrument Co., Fullerton, CA. 61 than doubled the average background counts. In most in- stances, the IRT count far exceeded the background count. The time to peak response (TTPR) was defined as the elapsed time (in minutes) of the highest (peak) count over the 15 minutes of recorded LECL. Because the peak responses to some agonists (OZ, FMLP, ZAS, bacteria) were prolonged, and did not occur within the 15 minute period of recorded LECL, data analyses for comparative studies were also based on the slope of LECL responses over time. The slope responses were calculated by using the IRT count (log transformed) as the first data point and the next four data points. Data analysis for differences in neutrophil responses between groups of agonists was by means of a computer based statis- tical program“1 for one way analysis of variance (ANOVA, OWA), assuming P50.05 as significant. Where there was sig- nificance in the ANOVA design, comparisons of means was by the least-significant-difference (LSD) testo. $22£1£191£1_2£_LEQL: The specificity of the LECL re- sponse to each agonist was investigated by the inclusion of SOD in the reaction vials, thereby defining the contribu- tion of superoxide anion to the generation of LECL. The percentage of SOD inhibition for each agonist was compared by means of one-way analysis of variance (OWA). Superoxide Assay: The production of superoxide anion was measured by the superoxide dismutase-inhibitable °SAS Insitute Inc., Cary, NC. 62 reduction of ferricytochrome C, as previously des- cribed.53'54 Briefly, each reaction tube contained 2 X 106 neutrophils, ferricytochrome C, and each agonist with or without SOD (total volume 1.0 ml). Two sets of paired tubes were used for each agonist tested and for the unstimu- lated control with the SOD-containing tubes serving as the reference blanks for the non-SOD-containing tubes. One tube contained only ferricytochrome C in HBSS. All tubes were placed in a shaking water bathJ for 30 minutes at 37 C. After incubation, the reaction was stopped by the addi- tion of SOD to the tubes without SOD, HBSS was added to each bringing the final volume to 2.0 ml, and the tubes were centrifuged at 1000 X g for 20 minutes. After centri- fugation 1.5 ml of the supernatant from each tube was pi- petted into disposable semi-micro cuvettesn and absor- bances were read in a double-beam spectrophotometerl, initially blanked using HBSS in the sample and reference positions. With HBSS as the reference, the ferricytochrome control was scanned (between 540 and 560 nm) with the peak absorbance at 550 nm recorded. The supernatant was reduced with sodium dithionate and rescanned. The peak optical den- sity was consistently less than 0.750 assuring linearity of results.54 All samples were read sequentially in a similar manner. The absorbance, after adding dithionate, was indicative of the total amount of ferricytochrome C pCole Parmer Instrument Co., Chicago, IL. 63 available for reduction. The net nanomoles of superoxide produced were obtained by subtracting the nM 02- produced by unstimulated controls from those produced by each agon- ist. The values for each set of two tubes were averaged. The mM extinction coefficient for a 1 cm light path Fe2 + Cytochrome C (reduced)-ferricytochrome C (oxidized) was assumed to be 21.1.54 Values were recorded as nM 02-/2 x lo6 cells/30 minutes. Analysis of data for signifi- cance (P50.05) between treatment groups used a OWA with LSD comparison of means. ec o t e WBQ_§gun£s: The effect of age on the total and differen- tial neutrophil counts from each calf was determined by means of a general linear model (GLM) one way ANOVA” with LSD comparisons of significant (P50.05) means. For these studies, the seven day period starting at the time of pur- chase was designated as week one (the initial week of accli- mation). Subsequent weeks were designated as week two and week three, respectively. The study was limited to a three- week period. Effect of Blood Handling: The effect of maintaining collected blood and isolated cells at room temperature or in ice was evaluated by paired T (PRT) testsm for both assays (Tables 1.1 & 1.2). Graphic depiction of these ex- periments is shown in Figures 1.1 and 1.2. These tables and graphs will follow "determination of agonist levels" 64 (Materials and Methods). Room temperature significantly enhanced LECL, whereas iced cells produced a more signifi- cant superoxide response. Since the production of super- oxide anion was enhanced and since its production has been postulated to have a major role in the generation of LECL all blood samples and isolated cells were routinely main- tained in ice. The maintenance of blood samples and iso- lated cells in ice also reduced the background counts (LECL) and the superoxide production of unstimulated con- trol cells to consistently low levels (data not shown). Detgrminatign_gf_gggn1§§_L§xg1§: In dose response stu- dies of LECL and superoxide production agonist concentra- tions which produced maximal stimulation of human neutro- 53'54 were used. The effects of phil oxidative metabolism different agonist concentrations were analyzed by use of OWA with comparison of means using the LSD test. Two con- centrations of agonists or treatments were analyzed by PRT tests. To determine the role of complement factors, immuno- globulins and host recognition factors as opsonins for zymo- san, heated, unheated, autologous and homologous sera were compared by paired T tests in each assay. Since the use of heated serum for opsonization of zymosan markedly dimin- ished neutrophil LECL and superoxide responses, over un- heated serum, the amounts (percentages) of inhibition in both assays were compared by a GLM one-way ANOVA with LSD comparison of means. For the best neutrOphil responses, 65 zymosan (10 mg/ml) was opsonized with unheated, autologous serum. This quantity of zymosan gave an approximate ratio of 25 particles per neutrophil.54 To ensure consistency of stimulation, dose responses were repeated for newly pre- pared batches of soluble agonists. For assessing LECL, latex particles, phorbol ester (PMA), calcium ionophore (CI) and opsonized zymosan (OZ) were used. Because latex particles were difficult to re- move from the supernatants of superoxide assays and ad- versely affected absorbance, 02 was the only particulate agonist used in the superoxide assays, in addition to the soluble agonists PMA and CI. The factors which were inves- tigated in determining agonist concentrations for LECL are summarized in Table 1.1 and graphically in Figure 1.1, and those for superoxide production are summarized in Table 2-2 and graphically in Figure 1.2. The concentrations selected were 50 ul of latex particles (LECL only), 200 ng of PMA, 10'5 M CI, and 50 ul of opsonized zymosan. To ensure that cells were exposed to identical amounts of particu- lates, 100 ul of 02 were used in the superoxide assay (con- taining 2 X 106 cells). Identical concentrations of sol- uble agonists were used in both LECL and superoxide assays. Analysis of data for LECL generation and superoxide produc- tion was by OWA with LSD comparison of means where signifi- cant (P50.05). t I div dua Calf V r a l t e L ang_§gpg;gxig§_firgggg§ign: To evaluate the effects of the 66 .Sflgaefl How H8» 68¢ 3.." 253m 5 mafia 5.9% new»? «meme—Hg no ~08!qu «353 39.5 msomoaofiun no Agvmgafig 5H3 pong gaeafifiou Buoy @3885 so $9.2 pounds 5.? 63.3890 gs Emma—HG pong «535 BA .HBoHuH—HE HH “£8353 8:92 some» a 2818” iguuHPusBuHfiHmuumEToH «g as. «Buss N2 wooed E m 3N8 m> sod Bo 386 an m «Ho 9 sub . . a mug 88 . o 84 .86 m S $me gHzammwNo 8: on 63 .SN 68 683 8 2: .mi 66 m 35 H38 Hosuofi as on .m.z 38 m 2: 8” .me an as :35 *mdmo 83 386 E m 3H8 80H 9 35.598» 88 $13!; 2 3% no 5358980 3.3, m .3131; mlllldlpul§o 8.826 5 duo why 0 4.." 0.38. 66A Figure 1.1. Graphic Representation of Experiments Performed in the Selection of Treatments and Agonist Concentrations for Use in Studies of Luminol-Enhanced Chemiluminescence (Table 1.1) by Bovine Neutrophils. Different bar patterns represent the means for each comparison group. Two agonists or treatments were analyzed by paired "T" tests. Three or more agonist concentrations were analyzed by one-way ANOVA. RT-neutrophils kept at room temperature; Iced-neutro- phils kept in crushed Ice. OZU-zymosan opsonized with unheated autologous serum; OZH-zymosan opsonized with heated autologous serum; AOZ-zymosan opsonized with autologous serum; HOZ-zymosan opsonized with homologous serum; Paired T test analyses of room temperature versus iced cells and A02 versus HOZ were performed on untransformed peak counts, and were log transformed for graphic purposes only. Horizontal bracket with a centered asterisk (*) enclose groups that were significantly different; horizontal bar with a subtitle NS enclose groups with no significant differences. a is. a? is... a voice» 0 53260 :3 w .09 one .33.. b O 67 _E\uc 2.2:. :3 bani .\ O 01601 oosomwnnwsrt‘l? (D ID d- '0 N °ojwdo). . poouoqug—lougw f‘ 68 .NA 8503 5 5.98% Edam new»? 6383?? uosu<\z «Erma—H.898 no 88:qu «853 £05 $60ng no 69.: msomoHQ—Ra fig pong cg: «85$ Suov Eamon Bugs .8 39 Bands 5? 8...?88 causes. Sax? om 5% 83880830 sax? m 5386 vogue «Hg ca .5386 gas 3% on 539533 «a; 3 Egan-cg “guns «£8ng 2 3mm» a. g 2992 EEO 3 so 6.: e as one 88 9 38 9 m8 9 on .mi n 28 H: 8H 9. 38 as om Bo 886 e .30 9 Bo 8s 28.0 a 38m 9 8m 3 so .886 m as o3 38 m> SN «\2 .mi N as o3 can 9 SN . a mug 88 .o e 38 gHsammeo 8: S .8 cos .9 08¢ ms 8H .3. e 35 89.8 Hoflofi . mfido BB 886 m 380 83 m> gang 88 $2493 2 9% Ho 83% 33> m .33 no mm on: you m8 o .38 .NA 0.35. 68A Figure 1.2. Graphic Representation of Experiments Performed in the Selection of Treatments and Agonist Concentrations for Use in Studies of Superoxide Pro- duction (Table 1.2) by Bovine Neutrophils. Different bar patterns represent the means for each comparison group. Two agonists or treatments were analyzed by paired ”T" tests; three or more agonist concentrations were analyzed by one-way ANOVA; ZlO-zymosan (10 mg/ml); ZZO-zymosan (20 mg/ml); Clio-opsonized zymosan (10 mg/ml); 50 ul-fifty microliters of opsonized zymosan (10 mg/ml); 100 ul-one hundred microliters of opsonized zymosan (10 mg/ml); OZS-opsonized zymosan (5 mg/ml); OZZO-Opson- ized zymosan (20 mg/ml). All other abbreviations identical to figure 1.1. Horizontal brackets with a centered asterisk (*) enclose treatment groups that were significantly different; horizontal brackets with the subtitle NS enclose groups with no signifi- cant differences. 69 .\ \\\\\\\\\\\\\\\\\\\:°~ 9\\\\\\\\\\‘ 2\\\\\\\\‘a L.‘é\".:° (n z 3.30;; 0 0 00/ I 0/ “E— - .Vo~ L—- .\\\\\\\\\\ {W \\\\\\\\\\\\\\\\\ [400 ’0 [:00/36; 0 Zymoom Cid Opoodzod Zymm Lot; \\‘««0 9V~° N’s 'E- \3} x\\\\\\\\\\‘ .9 .\\\\\\\\\\\\\\\\\\\\\\\ ~90 [2: ~00 .\\\\\\\\\\\\\\\\\\\\\\ \\\\ 4:. EW \0 ulw g 9"ng ”sum" 90 ac; 9 prong Calcium Ester launcher. ( ) 0 (No/u") Phorbol 70 variability among calves and the effect of age on neutro- phil responses, randomized completely blocked ANOVAs, using calves as a block and age as a treatment effect, were per- formed for each agonist and for each assay. Analyses for significant differences of the treatment effects (age), where significant (£50.05), was by LSD comparison of means. Since the actual studies were initiated during the second week after arrival, these data analyses were for weeks two, three, and four after arrival, respectively. The calves at those times were approximately 2.5-3, 3.5-4, and 4.545 weeks of age respectively. MW: A series of experiments were performed in which the LECL peak and slope responses and superoxide production of neutrophils stimulated by agonists were compared to the responses stim- ulated by Pasteurella haemolytica (PH). In these com- parative studies, live (PHL), dead (PHD), live-opsonized (PHLO), and dead-opsonized (PHDO) bacteria were used. The possible effects of formylated oligopeptides (FMLP) and com- plement factors [CSa (ZAS)] on the oxidative metabolism of bovine neutrophils were evaluated by their inclusion in these protocols. The quantities of bacterial suspensions used, 50 ul for LECL and 100 ul for superoxide assays gave an approximate ratio of 25 bacteria per neutrophil134 and was nearly identical to the particle to cell ratio obtained with 02.54 The concentration of FMLP (10'5 M) was that 71 indicated to induce maximal oxidative response in other spe- cies.54 The level of ZAS used has been previously de- fined136 for bovine neutrophils. Analysis of data for these studies was by OWA with LSD comparisons of means, where significant (Pg 0.05). RESULTS: Neutrophil isolation techniques consistently yielded >95 % neutrophils of >958 viability as determined by differ- ential counts and trypan blue dye exclusion tests. Contami- nating cells were principally lymphocytes with few plate- lets and red blood cells. Because our protocols were lim- ited to calves less than 5 weeks of age, there were no pro- blems with eosinophil contamination. There were signifi- cant decreases in total WBC counts and in the percentage of neutrophils between week one and week two, but there were no significant differences between week two and week three. (Table 1.3) The heating (56 C for one hour) of autologous serum prior to use in the opsonization of zymosan particles, com- pared to the use of unheated serum, produced significant de- creases in LECL and superoxide production responses (Tables 1.1 and 1.2, Figures 1.1 and 1.2). The percentage inhibi- tion of LECL produced by the destruction of complement fac- tors (heating of serum) was significantly greater than the percentage inhibition of superoxide production (Table 1.4). 72 .modwm an unobafififlmfiag Hana: .383 3.3.3 3 manage 36 H 8.3:. .88 3.2 H 2.3 mung gum? agfigm no: go gonnaamz ahgaam you“? cusp? «noaxfludoumzmouooanoflazag aggmoggflgfig 38 «gmogfiefiooflfiflzg A53 ggflgfiflgga mm...” H Hm.nn mmoo H nn.w g «32 I Lmz I Hm.n + mH.wn 3.0 + mN.b 95. £83 #NO.N H $.mv «mm.o H MN.m g 3 £4.33 mflaoo ".0on on '3]? Eu H Bo 8 8 m5. 9338 .nA «Home .modwm um ucBHufififiQtE Boos 3 3898 an 3&2... .86 H 2.8.: 65 ~13 H 2.2. mung guano aflnuoamaam no: @928 Effimz «again How-mac @9551. «noaxabdonmfimooooanougg aggmogggum+§5uo§ 38 «guoggofififig 9:8 gggfiflgga wN.n H Hm.nn mn.o ...I. mné g Luz I Lmz I .36 + mafia 3.0 + mNK 03». fl cNo.N .I... mode - *wmé H nmd g a 3% 4%.. H8 08 I _.I .0. o. 0.2%.. -1.“ 80.34.. .flco .. «5 9.83-0 44.. 542.3. 73 Latex particles stimulated the highest LECL responses in all instances, and these responses were often an indica- tor of the maximal neutrophil response for each day's exper- iments. The LECL peak responses to latex, PMA, CI, and 02 were all significantly different from each other. The times of initial responses to latex particles were consis- tently shortest, those to PMA and CI statistically similar, and those to OZ the longest with all three treatment groups differing significantly. The time to peak response was shortest when neutrophils were exposed to PMA, but the re- sponse time was not significantly different from neutro- phils exposed to latex. The peak response times for latex and PMA were significantly different from the responses to CI and to 02. The responses to CI and 02 were also signifi- cantly different (Table 1.5, Figure 1.3). Although the inclusion of superoxide dismutase (SOD) di- minished the LECL peak responses of neutrophils to soluble agonists (PMA and CI) more than for particulate agonists (latex and 02), there were no significant differences in the percentages of LECL-inhibition produced by SOD. (Table 1.6). Phorbol ester consistently induced the maximal produc- tion of superoxide anion with the responses differing sig- nificantly from those to calcium ionophore and opsonized zymosan, which were not significantly different (Table 1.7). “cabana? 82!. «muz .665 >958 B 3&3. 8.... H 36m... £9 cg 3.3.on mméa H 2&9... £8 §H 5:38 N: H ought. 35 H38 Hoauofi :d H 3693. U893 a I\+c I 54.3an 3838 “Eng .mnnz aggm H336 0.8 Ba minim. «Hana-«.8 3&8 Hugo magnum dd... «sumo—ER pong “80:998.“ 8.5.055 «Hound Honuonmusm «goats-cussing 93 «82¢ 2.3.80 .3 323.2...“ .5... ..\+ 5.8. 5 8.8 {3.0 H fim.¢H fimfl.o H Hh.¢ «5.0.0 H hm.N NO {—6.0 H ant—H. 3.0 H no.n {‘10 H ow.n U I imz I I Nm.o + 8.5 2.0 + Hh.N «mo.o + vH.¢ g £32 ev.o H de «do H omen «330 H med Xmas 74A Figure 1.3. Bovine Neutrophil Luminol-enhanced Chemiluminescence (LECL): Comparison of the Peak Responses, Initial Response Times, and Peak Response Times Induced by Soluble and Particulate Agonists. Analyses by one-way ANOVA (OWA) with comparisons of significant (P50.05) means by the least-signif- icant-difference (LSD) test. Each bar represents the mean of each treatment group, and the res- ponses induced by each agonist. *-Significant differences between treatment groups. The hori- zontal bracket and subtitle NS represents treat- ments that were not significantly different be- tween enclosed groups. N-number of experiments. N=35 I'\ Lb LI'U V's- rb N 4- D [semquII Emu asuodsau IBIIIuI a Latex E PMA B CI 5 oz v N — [Wd'JllJl 501 aouaosaquInIIwaLIg paouequm _I0LIIuIn'I 76 .56“ xmma can 8H5 3am: 5253 ”Egon“? hgmoHqung «30 Emma mumé 858.8 #33 “30 9.33 Tmé fig 382 «30 9.83 nlm.m 358.03 #33 £58 #353 unwaz .593 33 HR gag? HH “8% «egalmo “835% 8938.8 age—gang Eng Gong «EEH BHoHBuU 56o Hoflnmnsm “$9.4 x83 329.8 can; 3 flag... atmmmo .o Hoodo 0 fl 3 m2 H00 .0 m I NC «#wnmo .0 0a a .0 ¢ .fiw H0 m2 ”A m I HU m2 g m g «mm 94 m2 5 I g m2 ing .0 s. g owns ubouum fl god m 1% mammflwwlm .mnnz 88 En so a? an»? $355de 83% u8 Eu 589.. .355 unsung 85o .88 “5.83% HHESEEH? Emma—Eu owing 3.3.180.“ EHOHmDuHU :3va Hanson; Ens on\m:mo 93x «\INoifiuoflomma “ggbgflgmHgfifiB «sumé H mné “a «sand H 5H5 “a smmé H mafia «a 77 Randomized complete block analyses showed calf effects in the LECL peak responses to latex, CI, and OZ, but not in the response to PMA. There were significantly diminished responses between week 3 and week 4 with CI and 02. For superoxide production only 02 had a significant calf effect but there were no significant effect of age for any agonist. (Table 1.8) Comparisons of the LECL peak responses of neutrophils to agonists and to the Pasteurella organism, FMLP, and 2A8 are depicted in Figure 1.4. The magnitude of neutro- phil peak responses to latex, PMA, and CI were all signifi- cantly greater than to any of the preparations of Pasteur- ella organisms. Although differing significantly from the neutrophil responses to latex, PMA, and CI, the re- sponses to live opsonized Pasteurella organisms (PHLO) were statistically equal to the responses to 02. The opson- ization of live Pasteurella organisms significantly en- hanced the neutrophil LECL response over PHL, PHDO, and PHD. While the cell responses to OZ did not differ signi- ficantly from those to live Pasteurella (PHL) or to dead-opsonized Pasteurella (PHDO), they were signifi- cantly different from FMLP, ZAS, and dead Pasteurella organisms (PHD). Analysis of neutrophil slope response data (Figure 1.5) from the same series of protocols revealed significant 77A Figure 1.4. Bovine Neutrophil LECL: Comparisons of the Peak Responses Induced by Soluble and Particulate Agonists and Those Induced by Pasteurella haemoly- tica, Formyl-Methionyl-Leucyl-Phenylalanine (FMLP) and Zymosan-Activated Serum (ZAS). Analyses by OWA and LSD. LIX-latex particles; PMAPphorbol ester; CI-calcium ionophore; PHLO-live opsonized Pasteurella; OZ-opsonized zymosan; PHL-live Pasteurella; PHDO-dead opsonized Pasteurella; PHD-dead Pasteurella; each bar represents the mean of each treatment group with a standard error bar. Horizontal brackets with a cen- tered asterisk (*) enclose groups that are signifi- cantly different. Bracket with a subtitle NS enclose groups that are not significantly different. N-number of experiments. D II 2 F" En -gg F g é: I:g*"s * ‘ E .I‘ :3 I \g :3 mmmmmm D BBBBBBBBBBBBBBBBB 78A Figure 1.5. Bovine Neutrophil LECL: Comparisons of the Slope Responses Induced by Soluble and Particu- late Agonists and Those Induced by Pasteurella haemolytica, FMLP, and ZAS. Analyses by OWA and LSD. Abbreviations of agonist groups identical to figure 1.2. Each bar represents the mean of each treatment group with a standard error bar. Asterisk (*) centered over a horizontal bracket encloses groups which differ significantly. Bracket with a subtitle NS enclose groups that are not significantly different. N-number of experiments. D PHD ZAS FMLP N=IU Cl PHLU PHL oz {g {—" .— PM lllllllll 0000000 DDDDDDDDD 80 differences between the responses to latex, the twosome of PMA and CI, and the threesome of PHLO, PHL, and OZ. The latter two groups did not differ significantly within each respective group. The slope responses to PHLO differed significantly from those to PHDO, PHD, ZAS and FMLP. Superoxide production is depicted in Figure 1.6. The bovine neutrophil superoxide production responses to PMA and 02 were significantly different from each other and from the responses to CI, PHLO, and PHL. The responses to CI and PHLO were significantly different from those to PHDO, PHD, and ZAS. DISCUSSION: Neutrophils have come under intense biologi- cal scrutiny in the past decade, with the publication of over 25,000 scientific articles in the last five years alone. This interest has been not only in the definition of neutrophil cellular biology, but in the role of neutro- phils and neutrophil products in disease. A number of di- seases have been identified as neutrophil-mediated through neutrophil depletion studies in experimental ani- mals.128’129'182 Recently, Slocombe et.al129 performed neutrophil depletion studies in neonatal calves and defined the role of the neutrophil in the pathogenesis of lung in- jury in experimentally-induced pneumonic pasteurell- 0318.129 However, the definition of the neutrophil pro- duct(s) responsible for the characteristic gross and micro- scopic lesions, has been hampered by the complexity of 80A Figure 1.6. Bovine Neutrophil Superoxide Production: Comparison of Responses Induced by Soluble and Part- iculate Agonists and Those Induced by Pasteurella haemolytica, FMLP, and ZAS. Analyses by OWA and LSD. Abbreviations of agonist groups identical to figures 1.2 and 1.3. Each bar represents the mean of each treatment group with a standard error bar. Horizontal brackets with a cen- tered asterisk (*) enclose groups that are signifi- cantly different. Bracket with a subtitle NS enclose groups that are not significantly different. N-number of experiments. O %%%W PMA DZ Cl PHLU PHL PHDIJ PHD ZAS FMLP N=l3 NS ' I—'"—I a ‘_ §- [1- mmmmmmmmmmmmmmm NNNNNNNNNN °UIw DB/SIIBO 90 I xzxaPIXO-Iedns WN 82 interrelated in vivo pathways, necessitating in vitro studies for delineation. In the comparative studies of neutrophil responses to defined concentrations of agonists versus the responses to Pasteurella organisms, the magnitudes of neutrophil LECL (peak and slope) responses to latex, PMA, and CI were significantly greater than responses to 02 and live opson- ized Pasteurella organisms (PHLO). For superoxide pro- duction the responses to PMA and 02 were significantly high- er than to PHLO and CI. Whether the differences in the mag- nitudes of the LECL and superoxide responses are of biologi- cal importance is difficult to assess since the highest re- sponses were to non-biological entities. What is of major importance are the similarities of the neutrophil responses to PHDO and 02, since both particles are opsonized via the alternate pathway and because the par- ticle to cell ratios (25:1) were approximately equal.54’133 The differences in LECL and superoxide pro- duction responses stimulated by OZ and PHLO suggest: I) that differences in the mechanisms of induction of LECL and superoxide production by similar particulate agonists must exist, or 2) that difficulties exist in the attempted corre- lation of LECL and superoxide production data. Since the generation of LECL has been attributed to the production of reactive oxygen species such as singlet oxygen, hydroxyl ion, superoxide anion, hydrogen peroxide and hypohalous 83 ion, it is conceivable that the attempt to correlate the effects of five luminol-reactive entities with the produc- tion of one (superoxide) would be difficult even when induc- ed by similar or identical agonists. It is also conceiv- able that a single agonist could induce the production of a decidedly different spectrum of these compounds due to the influence of other, but presently undefined, factors. The similarities in the neutrophil responses to PHLO and 02 suggests that the mechanisms of neutrophil transduc- tion by each are similar, although 02 induces the produc- tion of more superoxide. Since studies of Chang et. al.133 have shown that the ingestion of Pasteurella haemolytica by bovine neutrophils leads to a rapid dimi- nution of LECL, it is possible that the pathogens may have altered the production of superoxide, thereby explaining the differences between 02 and PHLO in superoxide produc- tion. The bovine neutrophil responses to the different treat- ments of Pasteurella were incremental, although not ne- cessarily significantly different, with PHLO inducing the highest responses, followed in order by PHL, PHDO, and PHD. These data suggest: I) that there is a factor produced by live organisms that enhances their uptake by neutrophils, 2) that the effect of this factor is enhanced by opsoniza- tion, 3) that the uptake of dead bacteria by neutrophils is also facilitated by opsonins, and 4) that the phagocytosis 84 of opsonized particulates, nonpathogenic (OZ, PHDO) or pathogenic (PHLO) induces a substantial release of reactive oxygen species (LECL), including superoxide anion. Because it is not possible to accurately quantitate the release of reactive oxygen species in tissues, the postulation that these in vitro events may be related to the pulmonary injury seen in pneumonic pasteurellosis is hypothetical. While the factor(s) which induce these responses, other than opsonins, is/are presently undefined, it is apparently not a formylated oligopeptide, such as FMLP, since bovine neutrophils do not have FMLP receptorsgg, and since the oxidative metabolism of bovine neutrophils induced by expo- sure to FMLP was minimal. The role of complement factors in these responses was minimized because exogenous comple- ment (ZAS) failed to induce significant oxidative metabol- ism. Although complement activation can occur in vitro by means of the release of specific granule compo- nentslgs , there is a requirement for the presence of serum. Serum-free media was used in all of these experi- ments. The pathogenesis of neutrophil-mediated injury has been postulated to be the secretion and/or release of toxic reac- tive oxygen species formed as a consequence of hexose mono- phosphate (HMP) shunt activation, the secretion and/or re- lease of enzymes from cytoplasmic granules, and the produc- tion of arachidonic acid metabolites.1"3 The series of 85 events culminating in neutrophil activation is complex. A series of soluble and particulate agonists have been uti- lized in an attempt to delineate these processes. The in- teraction of surface membrane receptors with diffusible or particulate agents and membrane perturbation facilitating ionic transport may directly or indirectly induce intracel- lular enzyme systems, and/or stimulate the release or pro- duction of second messengers. These may then be important in the generation of reactive oxygen species, cell motil- ity, and secretion.6 The complexity of transduction mech- anisms is exemplified by the fact that some agonists may in- duce oxidative metabolism, others selective secretion, yet others motility or combinations of responses. There are phylogenetic differences in neutrophil metabolism, re- sponses, and enzyme components, making interspecies compari- sons difficult.98’99 There have been no published stu- dies which have explored the oxidative responses of bovine neutrophils to the spectrum of agonists used in these exper- iments. Agonists were selected for use in these experiments be- cause they stimulate neutrophil transduction by differing mechanisms. The cell response to latex particles has been reported to occur as a function of particle size and the surface potential properties of the particles.186 In our experiments in which the LECL responses of l X 106 86 neutrophils to four concentrations (25, 50, 75, and 100 ul/ml) of 0.8 micrometer latex particles were compared, there were no significant differences in cell responses. Mechanisms of transduction may therefore involve the ini- tial perturbation of surface glycoproteins and sialic acid residues, leading to ionic transport with secondary activa- tion of membrane-bound enzyme systems and perhaps the forma- tion of intracellular second messengers. The extreme rapid- ity of the response to latex, as compared to the particu- late 02, suggests that the initial cell responses were due to membrane disturbances with later responses due to par- ticle phagocytosis. The induction of the NADPH-oxidase complex, the “respi- ratory burst" enzyme, has been attributed to the intracel- lular activation of protein kinase C (PKC) with transloca- tion of PKC to the plasma membrane.21 Tumor promoting substances such as phorbol ester (PMA) have been shown to directly activate PKC, bypassing the cascade of intracellu- lar events normally involved in its activation.19'20 The activation of PKC, although reported to be calcium-depen- 2+ dent, occurs at "vanishingly low" Ca concentrations (10'8-10'1OM).20’21 In response to micromolar concen- trations of calcium, PKC and a Ca2+ -requiring thiol pro- teinase (calpain) become associated with the plasma mem- brane. Calpain activation converts PKC into a proteolytic- ally modified calcium and phospholipid-independent form 87 which is released into the cytosol where it has access to other, but presently undefined, protein sub- strates.21’25'27 The use of the compound A23187, also termed calcium ionophore (CI), facilitates the transport of the divalent cation Ca2+ into the cytosol where it has a major role as a second messengerf"6 Increases of intracellular Ca2+ may activate numerous calcium-dependent enzymes, including PKC, phospholipase C, phospholipase A2, calmodulin, and calpain, all of which have roles in the mechanisms of neut- rophil transduction. Cellular activation by the combina- tion of CI and PMA, each at lower than "usual” concentra- tions for neutrophil effects, has been shown to mimic "nor- mal" intracellular events.24 Particulate agonists, such as opsonized zymosan (02) and perhaps Pasteurella haemolytica, may activate the phosphoinositide cascade pathways, but stimulate the re- lease of superoxide by a mechanism distinct from PMA, in that the stimulus is phospholipase A2-mediated, calcium- dependent, and independent of PKC.50 The release of arachidonic acid is also calcium-mediated in OZ-stimulated cells but not in PMA treated cells.51 While the defin- itive mechanisms of OZ stimulation have not been defined, it has been suggested that the response may partially in- volve receptor~mediated events through interaction with sur- face CSa or Fc receptors.50 Since the stores of these 88 receptors have been localized in the membranes of the speci- fic granules, the addition of new surface membrane (and re- ceptors) has been demonstrated upon the fusion of specific granules with the cell surface.11'14 This process may also insert a cytochrome b component32'34, which resides within the membranes of specific granules, into the mem- brane-bound NADPH-oxidase complex thereby completing activa- tion and facilitating oxidative metabolism. Studies of neutrophil oxidative metabolism have cen- tered upon measures quantitating "respiratory burst” activ- ity including chemiluminescence (CL) and the production of superoxide anion. The generation of CL, by neutrophils has been attributed to the production of compounds such as superoxide anion, hydrogen peroxide, singlet oxygen, hydro- xyl ion, and hypohalous ions.1'53'57 The use of luminol, a cyclic hydrazide which is oxidized to an electronically excited intermediate state (aminophthalate ion) which emits light photons upon relaxation to a ground state, has been used to augment "native" CL and increase sensitiv- ity.58'6o Studies of generated LECL have been utilized in the delineation of cellular oxidative metabolism and metabolic or acquired opsonophagocytic dysfunction.60 The use of luminol also obviates the need for dark adapta- tion of vials and reagents and for the performance of the procedure under red (actinic) light. 89 Superoxide anion, produced by many enzyme systems, is formed either by the univalent reduction of oxygen or by the univalent oxidation of hydrogen peroxide.1’5 In neu- trophils, the induction of the NADPH-oxidase complex re- sults in the formation of 02- and H202 and leads to the activation and maintenance of the HMP shunt.1’5’48 Superoxide and its metabolites, generated within phagolyso- somes and released or secreted into the microenvironment has been postulated to have a primary role in the peroxi- dative attack on cell membranes, leading to cell injury or death.5’45’46 Both LECL and superoxide production protocols were char- acterized by extreme day-to-day variability in neutrophil responses. Although protocols for LECL and superoxide pro- duction used the same groups of isolated cells and the same concentrations of agonists, correlations in cell responses among agonists and assays were extremely difficult as ad- dressed previously. These often diametrical responses were partially attributed to: 1) individual variability in neu- trophil function, and 2) the quantities of superoxide anion induced by each agonist which contributed to the LECL re- sponse. Superoxide dismutase inhibited the neutrophil LECL responses to soluble agonists (PMA and CI) to a greater de- gree than to particulate agonists (latex and 02), although 90 there were no significant differences in the percentages of inhibition between agonists (table 1.6). The variability in cell responses described for LECL and superoxide production protocols was also apparent in the analyses for effects of age on neutrophil responses (Table 1.8). There were significant calf effects in neutro- phil LECL responses to latex, CI, and 02 and significant age effects in the LECL responses to CI and 02. These ana- lyses confirmed the hypothesis of the variability in the LECL responses of individual calves. There was a signifi- cant decrease in the LECL responses to CI and OZ between 3.5-4 weeks of age and 4.5-5 weeks of age. For superoxide production, only 02 had a significant calf effect, but none of the agonists had an age effect. The reasons for the di- minution in the LECL cell responses to CI at the ages stu- died are uncertain, but they could reflect alterations of Ca2+-dependent enzymes or altered Ca2+ transport lead- ing to decreased oxidative activity. The lack of similar alterations of CI-stimulated superoxide production suggests that this effect was not due to decreased superoxide produc- tion. Since neutrophil responses to 02 are mediated by a calcium-dependent enzyme, phospholipase A2, and are also calcium-dependentso, the alterations discussed above could also influence its responses. These findings suggest that individual variability and undefined effects of age in the responses of neutrophils to particulates could explain 91 increased susceptibility to bacteria, such as Pasteurella haemolytica, supporting the hypothesis of Hauser et. al.127 that age-related variations in neutrophil function may influence the susceptibility to disease. The response of neutrophils to particulates is partial- ly dependent upon opsonization, as shown by the responses to Pasteurella organisms and opsonized zymosan in our experiments, with serum complement factors and immunoglob- ulins having major roles.164 The use of heated serum for opsonization of particulates (OZ) produced a more signifi- cant degree of cell response inhibition in LECL than in pro- duction of superoxide (table 1.4), suggesting that comple- ment fragments (and receptors) are more important for the generation of LECL and that immunoglobulin fragments (Fc receptors) are more important for superoxide production. In calves, the increased susceptibility could be princip- ally due to lack of serum opsonic factors since neonatal calves are both hypogammaglobulinemic and hypocomplemen- temic at birth.164 While immunoglobulins are passively acquired through ingestion and intestinal absorption, com- plement factors are not, and normal (adult) levels are not reached until approximately six months of age.145 In calves, IgGl is preferentially absorbed over IgGZ by an ap- proximate ratio of 7:1.141 The principal opsonic immuno- globulin in cattle serum remains a subject of debate with IgGl, IgGZ, and IgM all purported to have the major 92 141'144 Although IgGl and IgCZ fix complement by roles. the classical pathway, only IgG2 fixes complement by the al- ternate pathway.141'143 IgGZ also facilitates neutrophil phagocytosis.143 If IgGZ is the principal opsonin, a se- lective IgG2 deficiency could explain the increased suscep- tibility of calves to gram-negative organisms. Since a portion of the host defense against gram-nega- tive organisms, such as Pasteurella, is dependent upon the alternate complement pathways, even adequate levels of immunoglobulins might not be protective. There have been few studies147 quantitating the levels of classical and alternate pathway components with increasing age in calves. No one has attempted to correlate the quantities of classi- cal and alternate pathway proteins with opsonophagocytic dysfunction. The use of adult and neonatal serum as opson— ins for neonatal neutrophils, combined with the dependence 2+ of the classical and alternate pathways on Ca and Mg2+ , respectively, could allow definition of the role of each component in host defense. The data obtained in these comparative protocols will provide baseline information for further investigations into the role of neutrophil oxidative responses to the Pasteurella organism and/or its metabolic byproducts. Recently, several papers have been published which describ- ed the use of LECL-inhibition as a means of assaying the quantities of P. haemolytica cytotoxin through effects 93 on bovine neutrophils.133'135 The extremes of LECL day- to-day variability, within individual animals, as well as between animals, casts doubt as to the validity and repro- ducibility of this assay. Although LECL was inhibited, there was no definitive proof that the neutrophils were actually killed. Other bacterial species, such as Haemo- philus somnus and Brucella abortus, may produce low molecular weight adenine and guanine nucleotides which may inhibit the ingestion of radiolabeled staphylococci and diminish iodination.]‘15'120 If similar compounds are pro- duced by Pasteurella haemolytica LECL might be inhibi- ted as well. Since the induction of neutrophil oxidative responses to particulate entitieszl’50 is mediated by dif- ferent intracellular pathways than are responses to solu- ble21 or receptor-mediatedls'17 agonists, perhaps the cytotoxin selectively blocks only these pathways, while others remain intact. In the attempted definition of disease, in vivo stu- dies are often complicated by the diversity of biological pathways, agonistic and antagonistic, which have major roles in the induction, prevention, exacerbation or modula- tion of tissue injury. In the attempts to define the patho- genesis of disease, particularly in man, animal models are extremely important. Future studies must of necessity in- volve the interpolation of in vitro results to in vivo events. The further definition of the role of 94 neutrophil products, and the attempted modulation of the factors responsible for injury, in pneumonic pasteurellosis may facilitate the understanding and possible treatment of fulminant gram-negative pneumonias in other species, includ- ing man, with obvious side benefits to the cattle industry. In summary we have: 1) defined methods for the use of luminol-enhanced chemiluminescence and superoxide produc- tion in investigations of the oxidative metabolism of bo- vine neutrophils; 2) established concentrations of soluble and particulate agonists that maximally stimulate the oxida- tive metabolism of bovine neutrophils; 3) defined parame- ters of blood handling and agonist preparation that may in- fluence the results of experiments; 4) determined that in- dividual variability and age of calves may influence the re- sponses to certain agonists; 5) determined that the re- sponses of bovine neutrophils to Pasteurella haemoly- ties are highly dependent upon the presence of opsonins and/or live organisms and that opsonized live, live, or dead opsonized organisms stimulate the production of sub- stantial amounts of reactive oxygen species, including superoxide anion, and 6) concluded that the production of these compounds may have a partial role in the lung injury which results from the interaction of neutrophils and Pas- teurella haemolytica. CHAPTER 2. THE EFFECTS OF ENDOTOXIN ON THE OXIDATIVE METABOLISM OF BOVINE NEUTROPHILS: COMPARATIVE EFFECTS OF PASTEURELLA HAEHOLYTICA AND ESCHERICHIA.COLI ENDOTOXINS INTRODUCTION: Bacterial lipopolysaccharide, also termed endotoxin, is derived from the cell wall of lysed gram negative bacterial organisms. The substance has a broad spectrum of biological activities affecting humoral and cellular host mediation systems and has been extensively reviewed.73 Endotoxin may trigger the activation of complement and coagulation pathways, in vitro or in vivo, and has profound effects on platelets, neutrophils, macrophages/monocytes, endothelial cells, mast cells and basophils.73 The interrelationships of these humoral and cellular pathways in the pathogenesis of disease are com- plex. Pasteurella haemolytica has a major role in the bo- vine respiratory disease (BRD) complex. This gram negative bacterium produces an endotoxin and cytotoxin during its log phase growth, both of which may be important in the pathogenesis of the pulmonary lesions which are hallmarks of the disease. Neutrophil depletion experiments have af- firmed the necessity for neutrophils in pneumonic pasteur- ellosis.129 Because endotoxins from gram negative 95 96 bacteria are similar in composition and biological effects, the potential role of endotoxin in the pathogenesis of pneu- monic pasteurellosis was investigated by comparing the effects of endotoxin (Escherichia coli OSS:B5) adminis- tered intravenously and intratracheally to anesthetized neonatal calves. While intravenous endotoxin was rapidly fatal, airway administration of identical quantities caused no significant physiologic or pathologic changes.188 The pathogenesis of neutrophil-mediated injury is thought to be related to the production of oxygen-derived free radicals (102, H202, 02-,OH'), release of enzymes from stores of intracytoplasmic granules and produc- tion of arachidonic acid metabolites (prostaglandins and leukotrienes).1'3 Results of in vitro studies have indicated that exposure of human neutrophils to trace amounts of endotoxin enhanced neutrophil oxidative metabo- lism, specifically the production of superoxide anion.76 The question posed in the present study was whether trace amounts of endotoxin had similar effects on bovine neutro- phils. The generation of luminol-enhanced chemilumines- cence (LECL) and production of superoxide by stimulated bo- vine neutrophils were used to investigate the effects of endotoxin on neutrophil oxidative metabolism. MATERIALS AND METHODS: Neonatal Holstein bull calves were purchased from a large local dairy herd. The calves upon arrival at MSU received an oral rotavirus and 97 coronavirus vaccine8 and intramuscular vitamin E/selen- iumb and vitamin ADEc injections upon arrival. The calves were examined twice daily with body temperature, pulse rate, respiratory rate, appetite, attitude, and ap- pearance of feces recorded. The calves were exercised once daily. Several blood samples were drawn during the initial week of acclimation for total and differential white blood cell (WBC) counts, which were used as an indicator of the health of the calf, and as a guide to the amounts of blood (and neutrophils) needed for completion of the protocols. Several aliquots of serum were harvested for use in the preparation of autologous opsonized zymosan (02). The calves were used for a maximum of 3 weeks. Blood samples were collected in syringes containing acetate-citrate-dextrose (ACD) anticoagulant, and neutro- phils were isolated by differential centrifugation and hypotonic lysis. The final isolates of cells were maintain- ed in calcium- and magnesium-free Hank's balanced salt solu- tions (HBSS) as previously described in chapter 1. Test tubes containing the isolated cells were maintained in crushed ice until used in the experiments. CHEMILUMINESCENCE: Luminol-enhanced chemiluminescence (LECL) was quantitated [in counts per minute (CPM)] in a Beckman LS 7500 liquid scintillation counterd programmed aCalf Guard, Norden Laboratories, Lincoln, NE. Bo-Se, Schering Corporation, Kenilworth, NJ. cVitamin ADE Injection, W.A. Butler Co., Columbus, OH. Beckman Instrument Company, Fullerton, Ca. 98 to count at 0.1 minute intervals for 15 minutes in an in-coincidence mode. A 70 uM (end point concentration) of luminol was used to enhance the generation of CL. A series of agonists, which stimulate neutrophil LECL by different mechanisms, were used including latex particles (LP), phor- bol ester (PMA), calcium ionophore (CI) and opsonized zymo- san (OZ). The methods of agonist preparation and quanti- ties used to stimulate neutrophil LECL and superoxide prod- uction were as described previously (Chapter 1). For analy- sis of data, the five data points (counts) for each minute of elapsed time were averaged and log transformed. Luminol- enhanced CL peak counts, slope responses, times of initial response and times of peak response were calculated as prev- iously described in Chapter 1 and recorded for statistical analyses. SUPEROXIDE PRODUCTION: The production of superoxide anion by stimulated neutrophils was quantitated by the superoxide dismutase-inhibitable reduction of ferricyto- chrome C53'54 as previously described (Chapter 1). The series of agonists used to induce the production of superox- ide consisted of PMA, CI, and 02. Unstimulated control cells were included in each assay. The amounts of superox- ide produced by the stimulated cells were calculated in nM of 02-/2 X 106 cells/30 minutes and were recorded for statistical analyses. 99 PREPARATION OF ENDOTOXIN: Phenol-extracted Escherichia coli 055:B5 endotoxin (EC:LPS) was obtained from commer- cially available sourcese. Phenol-extracted Pasteur- ella haemolytica endotoxin (PH:LPS) was generously sup- plied by Dr. Robert Walker, head of the bacteriology/myco- logy section of the Animal Health Diagnostic Laboratory at Michigan State University. The disposable capped polysty- rene test tubesf used in the preparation of both endo- toxins were those recommended by Cape Cod Associates, Woods Hole, MA. for the preparation and subsequent quantitation of endotoxin solutions, since they contained virtually no contaminating endotoxin. Two milligrams of PH:LPS and EC:LPS were diluted in 5 ml of endotoxin-free water. The test tubes were capped, and the preparations of PH:LPS and EC:LPS mixed by a vortex mixer at room temperature for one hour. To enhance solubility, the capped tubes were refri- gerated at 4 C for 24 hours and again mixed at room tempera- ture for 30 minutes. These were designated as stock solu- tions for both endotoxins. Dilutions, using endotoxin-free water, of each stock so- lution were made into polystyrene test tubes to an approxi- mate concentration of 1000 ng per ml. The tubes were cap- ped and 10 ml aliquots of each endotoxin solution were packed in dry ice and shipped to Cape Cod Associates in Woods Hole, MA. for endotoxin quantitation by turbidometric eSigma Chemical Company, St. Louis, MO. fFalcon Tube #2095, Falcon, Oxnard, CA. 100 Limulus amebocyte assay. An aliquot of prepared HBSS with Ca2+ 2+ and Mg , used in LECL and superoxide production protocols, was also submitted for quantitation of contami- nating endotoxin. Upon receipt of the quantitative report, final dilutions of each endotoxin stock solution were made to ensure quantitatively equal potency (1000 ng per ml). The sample of HBSS submitted was reported to contain 0.001 ng of endotoxin per ml. These preparations, designated "working dilutions" were stored in capped endotoxin-free borosilicate glass liquid scintillation vials at 4 C. This method of storage of the working dilutions was suggested by Cape Cod Associates as being able to preserve the stability of LPS preparations for as long as two years without signi- ficant loss of potency. The scintillation vials containing the working dilutions were mixed by vortex mixer at room temperature just prior to addition of the selected quanti- ties of endotoxin to the neutrophil suspensions. -ERIMENTAL DESIGN: It was assumed that the limit of maximal cell viability was 6 hours from the time of blood withdrawal, therefore all of the LECL and superoxide produc- tion experiments were completed within that time. The se- quence of agonists used to stimulate neutrophils was changed daily to ensure that the time which had elapsed after blood withdrawal did not influence the responses of the neutrophils. In this series of experiments, the re- sponses of unstimulated control cells exposed to the 101 selected concentrations of each endotoxin and those unex- posed to endotoxin were quantitated and included as an ana- lysis variable in each assay. Neutrophils and the selected endotoxin concentrations were incubated for 1 hour at 37 C. The four trace quantities of each endotoxin which were eval- uated included 10, 20, 50, and 100 ng/ml. These quantities represented the amounts to which the neutrophils were ex- posed during incubation. During the hour of incubation, LECL control assays (unexposed cells) were performed using the sequence of agonists selected for that day's experi- ments. After incubation with endotoxin for 1 hour, the en- dotoxinexposed neutrophils were used in LECL protocols (us- ing the same agonist sequence) and the superoxide proto- cols. Only one concentration of endotoxin could be evalu- ated in each day's experiments. The sequence of endotoxin concentrations investigated each week was randomized. A to- tal of 6 separate experiments were performed for each of the four concentrations of endotoxin. STATISTICAL DESIGN: For studies of the effects of PH:LPS and EC:LPS on neutrophil responses, superoxide production, LECL (peak and slope), and initial and peak response time data were analyzed by a computer based statistical program using three-way factorial analysis of variance (ANOVAs).g Factors evaluated included: 1) the variability in the re- sults of experiments which could be attributed to the 8SAS Institute Inc., Cary, NC. 102 responses (to all agonists) of neutrophils from individual calves (termed calf variability), 2) the differences in neu- trophil responses induced by the different agonists, 3) the differences in the responses of stimulated neutrophils in- duced by exposure to different endotoxin concentrations, and 4) the interactions of these factors. The least signi- ficant difference [LSD (T)]g test was used to compare means of the treatment group(s) when there was significance (P50.05) in the individual analysis factors (variables). To determine the overall effects of endotoxin exposure on neutrophil responses encompassing the responses to all agonists, a modified three-way factorial ANOVA was used in Which treatments (control, PH:LPS-exposed, EC:LPS-exposed) and responses to agonists were paralleled. In this design, interaction between treatment and agonist could not be de- termined. The analyses of PRT and IRT used only a modified three-way ANOVA for determination of overall significance, since it was assumed that endotoxin treatment and/or agon- ist exposure would induce the principal effects on neutro- phil responses. It was also assumed that neither a de- creased nor increased response time, alone, would play a significant role in the generation of LECL. RESULTS: Analyses of the neutrophil peak LECL response data (peak response 1, Table 2.1), indicated that there were no significant effects due to calves or to endotoxin levels, but there were significant differences in the peak 104 responses induced by different agonists, with significant interactions between calves and agonists, and between calves and endotoxin levels. Modified three-way analyses (peak response 2, Table 2.1) in the overall design revealed significant differences in neutrophil responses due to endo- toxin exposure with both endotoxins significantly decreas- ing cell responses. There were also significant effects due to agonist, but no effect of endotoxin level, and no interaction. Both endotoxins significantly reduced the LECL peak responses of neutrophils to PMA and CI (Figure 2.1). Analyses of the neutrophil slope responses (slope re- sponse 1, Table 2.1) showed no calf effect, a significant agonist effect, and no effect of endotoxin level, with no interaction. Modified-three way analysis (slope response 2, Table 2.1) showed no effect of endotoxin exposure, a sig- nificant effect due to agonist, and no effect of endotoxin level with no interaction. Both endotoxins significantly decreased the neutrophil slope response to PMA whereas only EC:LPS significantly decreased the response to CI over con- trol cells (Figure 2.2). With the exception of this latter effect, the peak responses and slope responses paralleled each other. When the initial response times (IRT) of stimulated neu- trophils were analyzed, there was no significant effects of endotoxin incubation or endotoxin concentrations, but there 104A Figure 2.1. Comparison of the LECL Peak Responses of Bovine Neutrophils Exposed to Trace Quantities of Pasteurella haemolytica Endotoxin (PH:LPS) and Escherichia coli Endotoxin (EC:LPS) upon Stim- ulation by Soluble and Particulate Agonists. Analyses by three-way factorial ANOVA with comparison of significant (P50.05) means by the least-signifi- cant-difference (LSD) Test. Each bar represents the means of each treatment group with a standard error bar. PMA-phorbol ester; CI-calcium ionophore; OZ-op- sonized zymosan; UC—unstimulated cells; All compari- sons were made to control cells. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significantly from control cells. Hori- zontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. 105 E ED E\\\\\\\\\\\\\\\\\\\\< .\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\\ ,/ ;/' ,,,,,,, EIo-I soueosaquInIIwaLIo peouequa_Ioqum'I US UZ Cl PMA L/ATEX '4 Control i PH Endotoxin 8 EC Endotoxin N =24 105A Figure 2.2. Comparison of the LECL Slope Responses of Bovine Neutrophils Exposed to Trace Quantities of PH:LPS and EC:LPS upon Stimulation by Soluble and Particulate Agonists. Analyses by three-way factorial ANOVA and LSD. Each bar represents the means of each treatment group with a standard error bar. Abbreviations of each treatment group identical to Figure 2.1. All comparisons were made to control cells. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. 106 rm U6... 000000000000000000 000000000000000000 l2 QZZQZQZZ §§§§§§§ DZ a EC Endotoxin 0000000000000000 00000000000000000 .\\\\\\\\\\\\\\\\\\\\\\\ PMA LATEX a Control E PH Endotoxin /’ IQZVl' V§§$§§§$§§§§~ 010- 0.00- 1 I: N D 070— 0.00 - 0.50 -— 0.40 4 0.30 ~ eInqu Jad sIunog moo-1 sdoIS N=Z4 107 were significant differences in the IRT between agonists. There were no interactions. Only the unstimulated cells exposed to PH and EC endotoxins had prolonged IRT, but none of the responses to other agonists differed significantly from control cells (Table 2.1, Figure 2.3). Data analysis for endotoxin effects on neutrophil peak response times (PRT) revealed no significance due to endo- toxin incubation or endotoxin concentration and no interac- tion. Only the agonists induced a significant effect, with CI-stimulated cells showing prolonged PRT after incubation with both endotoxins (Table 2.1, Figure 2.3). The results of three-way factorial ANOVA analyses of superoxide production data (superoxide 1, Table 2.1) showed significant differences in the responses of neutrophils from individual calves, and significance due to agonists, but none due to endotoxin concentrations. There was, how- ever, a significant interaction between calves and endotox- in level. The neutrophils from calf A2 produced signifi- cantly more superoxide than those from calves Al and A3. Modified three-way analyses (superoxide 2, Table 2.1) revealed a significant effect of both endotoxins on neutro- phil superoxide production, but there was no effect of endo- toxin concentration. The different agonists stimulated the production of significantly different quantities of superox- ide after endotoxin exposure. The superoxide production of PMA-stimulated cells exposed to both endotoxins increased 107A Figure 2.3. Comparison of the LECL Initial and Peak Response Time Responses of Bovine Neutrophils Exposed to Trace Quantities of PH:LPS and EC:LPS upon Stimu- lation by Soluble and Particulate Agonists. Analyses by three-way factorial ANOVA and LSD. Each bar represents the means of each treatment group with a standard error bar. LTX-latex particles; all other abbreviations for treatment groups identical to Figures 2.1 and 2.2. All comparisons were made to control cells. Horizonal brackets with a centered asterisk (*) enclose treatment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. 108 ?I/lT/I/4'/A \\\\\\\\\‘ 0.00.0000...OOOOOOOOOOOOO... 060.000.060.000...0.00.000... \\\\\\\\\\\\\\\\\\\\\\\\\\\\\ OZ UC NS Cl ..... O'VVV' LTX PMA 8 EC endotoxin T T I 1 I I I vvvvvvvvvv Cl m '9'o'o'o'o'o'o.o.o'o '0'... LTX PMA 2 Control I I j I I I ID V- CU N - D [semquI] emIl esuodseu lalllul N=Z4 3 PH endotoxin 109 significantly, but the responses of EC:LPS-exposed cells were also significantly higher than those exposed to PH:LPS. The CI- and OZ-stimulated responses of cells exposed to both endotoxins also increased significantly, but there were no significant differences between the effects of the endotoxins (Figure 2.4). DISCUSSION AND CONCLUSIONS: The responses of neutrophils to endotoxin, in vitro, have been reviewed73’85, and the specific responses to trace quantities of endotoxin (< 100 ng) investigated.76’77'93 These data from my studies represent the first reports of the effects of trace endo- toxin concentrations on the oxidative metabolism of bovine neutrophils and the first comparison of the effects of EC:LPS and PH:LPS. Since the responses of neutrophils re- ported previously76’93 required the incubation of the cells with endotoxin for a minimum of 30 minutes and gave a maximal effect after incubation for 60 minutes, only the latter incubation time was used in our protocols. I chose to use the series of agonists and agonist concentrations previously defined (Chapter 1) to determine the effects of endotoxin exposure. Previous reports76’93 had shown that exposure of neutrophils to trace quantities of endotoxin caused significant increases in superoxide production when the neutrophils were stimulated with PMA and 02. In addi- tion to PMA and 02, my experiments also used latex and CI as agonists to determine the effects of endotoxin exposure. 109A Figure 2.4. Comparison of the Superoxide Production of Bovine Neutrophils Exposed to Trace Quantities of PH:LPS and EC:LPS upon Stimulation by Soluble and Particulate Agonists. Analyses by three-way factorial ANOVA and LSD. Each bar represents the means of each treatment group with a standard error bar. Abbreviations for treatment groups identical to figures 2.1, 2.2 and 2.3. All comparisons were made to control cells. Horizontal brackets with a centered asterisk (*) enclose treat- ment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. 110 =24 “‘0'0 NS UC ooooooooooooo oooooo 09900.. 9000900000... 0000000000000 900000900000. 900000000000. * OOOOOOOOOOOOOOOOOO O O O O O 0 O O O O O O 0 O O 0 O 0 o O O O O O O O Q O O O O O O O O O O O O O O O O O O O O O O O O O O 0 § 0 O O O O O 0 O O O O O O O O O O O O O O O O O O O O O O O O O O O O V ............................................ ..'.'...'.'.C.‘. V ‘ WOOOOOOOOOOOOOO00.009000900090009O OOOCQOOOOOOOOQO0.00.0.9....0.0.90 OQOOOOOOOOOOOOOOOOOOOOOOOOOOOOOQO 9000000000.900...00.000.09.000... ooooooooooooooooooooooooooooooooo * W PMA '4. Control 8 PH Endotoxin 8 EC Endotoxin N I I I j l I II N V‘ LDNmV’D 0'.) N —-— 20- I U3 N W nee/Etna“-I olxzxepIxmadns WN 111 Since these agonists all trigger neutrophil oxidative meta- bolism by different mechanisms, I hoped to selectively de~ termine which pathway(s) might be enhanced by exposure to endotoxin. Since endotoxin increases superoxide production to selected agonists, I reasoned that the compound might induce similar effects on generation of LECL by stimulated neutrophils. Therefore, my investigation was expanded to evaluate the effects of endotoxin on LECL. Experiments to compare the effects of EC:LPS and PH:LPS on bovine neutro- phil oxidative metabolism were designed to: 1) resolve the premise that the biological effects of endotoxins vary little between bacterial species; 2) determine whether the effects of PH:LPS were specific for bovine neutrophils; 3) explore the hypothesis that exposure of bovine neutrophils to trace quantities of LPS might "prime" the cells to pro- duce increased quantities of luminol-reactive entities (re- active oxygen species) including superoxide anion, after ex- posure to agonists; 4) determine whether these endotoxin ef- fects might be localized to a specific pathway or mechanism of neutrophil activation based on agonist responses; and 5) determine whether these effects might have a role in the pathogenesis of pneumonic pasteurellosis. To answer these questions I used techniques and ago- nists defined in Chapter 1 to investigate bovine neutrophil oxidative metabolism. These agonists were selected because they activate neutrophils by different mechanisms, as 112 previously discussed. Latex particles activate neutrophils as a function of the size and ionic properties of the par- ticles187 , with membrane perturbation of surface sialic acid residues leading to ionic transport as the initial ba- sis of responseé, followed sequentially by responses acti- vated by particle phagocytosis.186 Phorbol ester (PMA) activates neutrophils by direct stimulation of protein kin- ase C, an enzyme intimately involved in the activation of the "respiratory burst" enzyme, NADPH-oxidase, as well as in other sequential cellular functions.19'21 Calcium ion- ophore (CI) induces neutrophil activation by the transport of extracellular Ca2+ into the cytosol where it has a ma- jor role as a second messenger.4'6 The increase in intra- 2+ cellular Ca may trigger the activation of several cyto- solic and membrane-bound Ca2+ -dependent enzymes known to be important in cellular transduction. Opsonized zymosan activates neutrophils by mechanisms dependent on the activa- tion of phospholipase-A2 and the presence of cytosolic calcium.50 While the use of 02 has been suggested to ini- tiate the polyphosphoinositide cascade, the activation of cells by 02 is independent of protein kinase 0.50 Experiments described in chapter 1, in which the oxida- tive responses of bovine neutrophils were studied, revealed marked day-to-day variability in the neutrophil responses of individual calves as well as differences between indivi- dual calves. To assess the possible effects of variability 113 due to calf factors, three-way factorial ANOVAs were used to examine the responses of neutrophils. In these anal- yses, the interaction between calf factors and other vari- ables could be determined. In the protocols comparing PH:LPS and EC:LPS, only the analyses of superoxide produc- tion data (table 2.1) had significant differences in neutro- phil responses due to the calf from which the neutrophils originated. Only the analyses of data for LECL peak re- sponse and superoxide production revealed significant inter- actions of calf factors with other factors. The ”priming" phenomenon has been described as the en- hanced production of superoxide anion after incubation with trace levels of endotoxin (< 100 ng).76 In my studies I examined the effects of similar trace levels of endotoxin on LECL generation and superoxide production of bovine neu- trophils. The studies of the effects of endotoxin on LECL examined several neutrophil function parameters including peak response, slope response, times of initial response, and times of peak response. Examples of "priming" which might be anticipated using these parameters of neutrophil responses should hypothetically include: 1) an increase in the neutrophil peak response; 2) an increased slope re- sponse; 3) a decreased or faster time of initial response; and 4) an earlier time of peak response. For superoxide, production "priming" should enhance the production of super- oxide anion. If the hypotheses that were supported by the 114 increased production of superoxide by human neutrophils after stimulation by PMA and OZ are also true for bovine neutrophils, then similar changes (as above) should occur in the LECL parameters and superoxide production after stimulation by these agonists. The molecular basis of the "priming” effect was sug- gested to be: 1) alterations of the plasma membrane so that contact or binding of stimuli might be greater; 2) that events coupling stimulus-receptor interaction leading to activation of the NADPH-oxidase might activate more en- zyme; or 3) the conformation of or placement of the enzyme in the membrane might be changed so that it expresses higher activity.76 Since it has been shown that neutro- phils have the capability to metabolize endotoxin, it was postulated that the lipid A fractions might be metabolized to form fatty acids, which can induce superoxide release, and directly activate protein kinase C (PKC).76 It was suggested that the time and temperature requirements were consistent with the formation of a metabolite which might directly activate PKC or stimulate increased polyphospho- inositide metabolism and turnover indirectly activating ch.2° There was a significant effect of endotoxin-exposure, on LECL peak responses, seen with both endotoxins but no effect of endotoxin concentration. Both endotoxins signif- icantly reduced the peak responses to PMA and CI but had no 115 significant effect on the neutrophil responses to latex and to OZ. The slope responses paralleled the peak responses in that the responses to PMA and CI were also significantly reduced, but for the latter only the response to EC:LPS was significantly reduced from control cells. In support of the decreased peak and slope responses to C1, the time of peak response was significantly prolonged. In sharp con- trast to the LECL response data, exposure of neutrophils to both endotoxins induced significant enhancement of superox- ide production after stimulation by PMA, CI, and OZ. The interpretation of these diametrical differences in neutro- phil responses to the same agonists is difficult. Under the strict definition of "priming", only the superoxide pro- duction responses would fulfill the criteria discussed pre- viously. The inability to correlate the results of the LECL gen- eration and superoxide production assays has been addressed in Chapter 1. This is in part due to variable quantities of superoxide anion and other reactive oxygen species pro- duced by each agonist, and the relative contributions of each entity to the LECL response. The superoxide anion was shown to be responsible for a larger part of the LECL peak responses when stimulated by soluble agonists (PMA, CI) than those when stimulated by particulate agonists (latex, OZ). Since the production of superoxide in response to sol- uble agonists is enhanced by endotoxin exposure, the 116 production of other luminol-reactive entities such as 'OH, 102, H202, and HOCL' must be markedly in- hibited for peak LECL to drop so significantly. Since the catabolism or disposal of superoxide anion is mediated prin- cipally through the activity of cytosolic superoxide dismu- tase, decreased activity would decrease the amounts of H202 formed for disposal by other enzyme systems, such as catalase, the myeloperoxidase-halide system, and the glu- tathione peroxidase system. If there are increased quanti- ties of hydrogen peroxide produced, an inhibition of the myeloperoxidase enzyme might diminish the production of hypohalous radicals and singlet oxygen, thereby decreasing LECL. Another possible explanation is that the glutathione system might catabolize the formed hydrogen peroxide at an increased rate. Since it has been shown that bovine neutro- phils, when stimulated, have a marked activation of the hex- ose monophosphate shunt, and since the production of reduc- ing equivalents necessary for activation of the respiratory burst enzyme NADPH-oxidase are intimately tied to the gluta- thione system and the HMP shunt, it is possible that this might explain not only the increased superoxide production but also the decreased availability of hydrogen peroxide for use by other enzymes. Since the mechanisms of induction of neutrophil oxida- tive metabolism by particulates follow different intracell- ular pathways than for soluble agonist321'50, it is 117 possible that there is also a selective enhancement of the pathways for superoxide production and selective inhibition of others involved in the generation of LECL by particu- lates. Since there was enhancement of superoxide produc- tion after exposure to endotoxin and stimulation by 02, yet no significant increase in LECL, this suggests that the lack of correlation between the two assays might be ex- plained by: 1) the relative contribution of superoxide an- ion to the LECL response to particulates; and/or 2) a less significant inhibition of the production of other luminol- reactive entities, perhaps related to the different path- ways of neutrophil transduction used by particulate agon- ists. The decrease in LECL responses and enhancement of super- oxide production, after endotoxin exposure, to stimulation by PMA suggests three hypotheses: 1) that either PKC it- self and/or its substrates are altered by endotoxin; 2) that changes in membrane permeability allow less penetra- tion of PMA into the cytosol; or 3) that endotoxin alters enzymes distal to PKC which are necessary for the selective induction of oxidative metabolism. Hypotheses for the decreased LECL responses and enhance- ment of superoxide production, after endotoxin exposure, to CI must center upon: 1) altered Ca2+ transport, either reduced or enhanced, perhaps due to altered plasma membrane permeability; or 2) alteration of intracytoplasmic Ca2+ 118 dependent enzymes including PKC, calpain, phospholipase A2, phospholipase C, and calmodulin, perhaps selectively affecting production of other second messengers. Hypotheses for the the increase in superoxide produc- tion, after endotoxin exposure, to OZ include: 1) altera— tion of surface enzymes facilitating particle/surface inter- action; 2) up-regulation of surface complement or immuno- globulin receptors; or 3) selective enhancement of enzymes (phospholipase-A2) involved in the cell response to parti- culates. These postulates detail not only the complexity of neutrophil transduction mechanisms but also the diffi- culties in correlating the neutrophil responses between assays, and between species. In conclusion, these studies have determined: 1) that the exposure of neutrophils to endotoxin has profound ef- fects on neutrophil oxidative metabolism but results are dependent upon the assay procedure, the agonist(s) used, and the species from which the neutrophils were obtained; 2) that there were no significant differences in the re- sponses of bovine neutrophils after incubation with PH:LPS or EC:LPS, therefore ruling out a species-specific effect; 3) that there were no significant differences in the ef- fects of four different concentrations of endotoxin on the responses of exposed neutrophils; and 4) that these data support the contention that endotoxins vary little in their biological effects, differing only in their potency. In 119 the latter conclusion, potential variability between endo- toxins was eliminated by accurate quantitation of endotoxin concentrations. These data also support the principal hypo- thesis that exposure of (bovine) neutrophils to trace con- centrations (10-100 ng) of endotoxin may "prime" them to produce increased quantities of superoxide anion after stim- ulation by PMA, OZ, and additionally by C1 and that this re- sponse was irrespective of the species (PH or EC) from which the endotoxin originated. Therefore, the presence of small quantities of endo- toxin released by dying gram negative bacteria, such as Pasteurella haemolytica, may facilitate pulmonary in- jury via release of superoxide anion from "primed” bovine neutrophils. Since complement activation did not play a role in these in vitro studies (serum-free media), this "priming" effect was not due to the indirect action of com- plement factors. It is difficult to equate the results of my studies with those of Confer and Simons159 who used vastly differ- ent concentrations of Pasteurella haemolytica LPS to determine the in vitro effects on bovine neutrophils. They concluded: 1) that PH:LPS was not directly toxic to neutrophils; 2) that PH:LPS had little effect on the random migration of bovine neutrophils; 3) that exposure to PH:LPS enhanced the NBT reduction of exposed cells; 4) that low 120 (2.5 ug/lO6 cells) and high (65 ug/lO6 cells) concentra- tions decreased the phagocytosis of labeled Staphylococ- cus aureus; and 5) that moderate quantities (5-25 ug- /106 cells) increased the phagocytosis of labeled Staph- ylococcus aureus. While trypan blue due exclusion tests were not performed in our experiments after endotoxin incu- bation, the enhancement of superoxide production suggests that the neutrophil possessed fully functional metabolic pathways. Since nitroblue tetrazolium reduction is a quan- titative measure of superoxide production, the enhanced NBT reduction reported would correlate well with our results of enhanced production of superoxide anion. Since we did not perform random migration studies nor radiolabeled staphylo- coccal ingestion studies, no correlations are possible be- tween the results of my and their experiments. While their findings that staphylococcal ingestion was enhanced by "mo- derate" quantities of PH:LPS again provide no correlates to my studies, their data suggest that specific levels of endo- toxin facilitate the phagocytosis of bacteria. Whether this response is specific to staphylococci or whether simi- lar responses to Pasteurella would also be seen remains to be investigated. There have been no other reports in the veterinary literature which have reported the in vitro effects of PH:LPS on the oxidative metabolism of bovine neutrophils. 121 Future studies should involve the exploration of the effect of endotoxins on the cell responses to the Pasteur- ella organism in vitro and in vivo with the for- mer encompassing studies of oxidative metabolism and enzyme release and the latter the airway administration of endo- toxin followed by variable numbers of Pasteurella or- ganisms. CHAPTER 3: ENZYME RELEASE BY BOVINE NEUTROPHILS INTRODUCTION: The pathogenesis of neutrophil-mediated in- jury involves three major mechanisms: the release of reac- tive oxygen species, the secretion or release of enzymes from cytoplasmic granules, and the production of arachido- nic acid metabolites.1'3 The importance of neutrophils in the pathogenesis of pneumonic pasteurellosis of cattle has been identified by neutrophil depletion of neonatal calves129 , which prevented lung injury after intratra- cheal inoculation of virulent Pasteurella organisms. Although many of the mechanisms of neutrophil activation leading to tissue injury are understood, there is no uni- fying theory as to which neutrophil product(s) is/are re- sponsible. Cyclooxygenase inhibitors and polyethylene gly- col (PEG) bound catalase have not lessened the severity of experimentally-induced Pasteurella pneumonia in calves. (unreported data) Ruminant (bovine) neutrophils are unique, and possess three distinct cytoplasmic granule subtypes which may be differentiated ultrastructurally during myelopoesis by means of peroxidase reactivity.100'101 The biochemical properties of bovine neutrophils have been delineated99 122 123 and phylogenetic comparisons have been made to other spe- cies.98 When compared to human neutrophils, bovine neu- trophils lack lysozyme, and are relatively deficient in myeloperoxidase and in primary (azurophil) granule compo- nent enzymes. They have a large component of secondary (specific) granules and associated enzymes. Since bovine neutrophils are also relatively deficient in catalase, the pathways of hydrogen peroxide (H202) catabolism are considered to be unique in comparison to neutrophils of other species.99 To my knowledge there have been no previous studies which have explored the selective release of granule-associ- ated or cytoplasmic enzymes by bovine neutrophils, either in vivo, or in vitro. The relationships of these potentially damaging neutrophil products to fulminant pneu- monic pasteurellosis are presently undefined. The purpose of this research was to characterize the magnitude of en- zyme release by activated bovine neutrophils: first, to a series of defined agonists and subsequently to the Pas- teurella organism itself. Marker enzymes for neutrophil cytoplasmic granules have been previously defined and vary among species. For bovine neutrophils, beta-glucuronidase (BC) is the marker for azur- ophil granules, and vitamin B-12-binding protein (B12BP) is the marker for specific granules.99 Lactate dehydrogen- ase (LDH), a cytosolic enzyme, is frequently used as an 124 indicator of cell death or lysis associated with increased membrane permeability.53’54 MATERIALS AND METHODS: Experimental calves were acquired, acclimated, and maintained as previously des- cribed. Blood samples were obtained, and neutrophils were isolated by previously described methods. All reagents, analytical grade, were obtained from Sigmaa, and were pre- pared as previously described. To provide a basis of com- parison for future studies, a series of experimental proto- cols were designed to explore bovine neutrophil enzyme re- lease in response to the soluble agonists phorbol ester (PMA) and calcium ionophore (CI), and to the particulate agonist opsonized zymosan (OZ). Agonist concentrations used in these protocols were determined from those sug- gested for use in similar experiments with human neutro- phils53’54 and which were defined further in experiments exploring the oxidative metabolism of bovine neutrophils (Chapter 1). The concentrations used in these enzyme re- lease protocols were: PMA, 200 ng/ml, CI 10"5 M, and 50 microliters of autologous opsonized zymosan per 2 X 106 neutrophils. For comparative studies, suspensions of Pas- teurella haemolytica organisms (PHL, PHLO, PHD, PHDO) were prepared as previously described (Chapter 1) after be- ing adjusted to an optical density of 1.60 at 541 nm in a double beam spectrophotometerb. Preparations of 8Sigma Chemical Company, St. Louis, MO. bShimadzu UV-260, Shimadzu Instruments Ltd., Tokyo, Japan 125 organisms were maintained at room temperature. For stimula- tion 50 ul of each preparation of bacteria was used per each 2 X 106 cells, to give an approximate ratio of 25 bacteria per neutrophil.133 This particle/cell ratio was nearly identical to that of 02.54 Because of the low levels of BC and LDH in bovine neu- trophils, 10 X 106 PMNs were used to ensure detectable and reproducible quantities of enzymes (data not shown). For quantitation of B12BP, it was only necessary to use 2 X 106 cells. All experiments included positive controls, consisting of equal numbers of cells lysed with 1% tri- ton-Xa, and negative controls. Test tubes (10 X 75 mm) containing PMNs (2 x 106 for BIZBP, and 10 X 106 for LDH and BG), Hanks balanced salt solution with Ca2+ and Mg2+ (HBSS), and the selected levels of agonists, includ- ing positive and negative controls, were placed in a shak- ing water bath at 37 C for 30 minutes. After agitation, the tubes were centrifuged at 1000 X g for 20 minutes at 4 C. Tubes containing the supernatants were placed in crushed ice until the individual assays were performed. Beta-glucuronidase release was determined by the hydro- lysis of the chromogenic substrate phenolphthalein glucuron- ide as previously described53'54, by using an assay kit obtained from Sigma.8 Aliquots of supernatants from the degranulation assay were incubated in acetate buffer and phenolphthalein glucuronic acid for 18 hours at 37 C. 126 Linear regression was performed on absorbance data derived from a series of standard samples, was used to obtain a slope and intercept for quantitation of unknown samples. Post-incubation, glycine buffer was added for color devel- opment. A double beam spectrophotometer, zeroed at 540 nm with acetate buffer, was used to sequentially read all samples. Values for glucuronic acid controls were sub- tracted from all other readings. The amounts of BC re- leased in response to the agonists were quantitated as micrograms of phenolphthalein released per 10 X 106 cells/ 30 minutes. For determination of LDH activity, supernatants were added to a prewarmed (37 C) substrate kita in microcuv- ettesc and mixed, and after a 30 second delay the change in absorbance was followed kinetically at 37 C and 340 nm in a double beam spectrophotometer.b The changes in ab- sorbance which occurred at five consecutive 1 minute inter- vals were recorded. The data were averaged and used to cal- culate the enzyme release in Wacker units. Samples of ab- normal serum, generously supplied by Sigmaa, were used to ensure the accuracy of the assay. It was determined that the values (in Wacker units) of the positive controls and the abnormal controls were nearly identical and had excel- lent reproducibility. subsequently, only positive controls were performed in each assay. cCole Parmer Instrument Co., Chicago, IL. 127 For determination of B12BP activity54 , cyanocobalamin 57Co-vitamin B12 (12 uCi/microgram) was obtained from Amershamd, diluted in normal saline to a concentration of 5 ng/ml, divided into aliquots, and stored at 4 C in 0.6 m1 Eppendorf microcentrifuge tubesc. For calculation of spe- cific activity, the radioimmunoassay included total activ- ity and charcoal control (unbound) samples in all assays. Duplicate samples were prepared for all agonists. Sequen- tially, saline, supernatant, and 1000 pg of 57Co-vitamin B12 were added to Eppendorf microcentrifuge tubes (1.6 mlc) and mixed using a vortex mixer for 10 seconds. Af- ter mixing, 0.4 ml of an equal volume of 5% Norit charcoal and 1% bovine serum albumin (fraction Va) was added, and the tubes were again mixed with a vortex mixer. After mix- ing, the tubes were centifuged for 2 minutes at 12,000 RPM in a centrifuge with a microcentrifuge tube rotor. Equal amounts of supernatant (0.8 ml) were removed from each tube, taking care not to disturb the sedimented charcoal, placed into 10 X 75 mm polypropylene test tubes, and capped. Sample activity (CPM) was read in a Searle Analy- tic 1185-R gamma counterx. Counts derived from exposure of neutrophils to each agonist were recorded in three chan- nels. The readings from the three channels for each of the duplicate samples were averaged to calculate B12BP release. aAmersham Corp., Arlington Heights, IL. eSearle Analytic, Inc., Des Plaines, IL. 128 Values were recorded as pg vitamin B12 bound/2 X 106 neutrophils. The values for positive controls in each assay were con- sidered to be maximal possible release of enzymes. The max- imal values recorded for each agonist were used to deter- mine the percentage (%) release induced by exposure to each agonist. EXPERIMENTAL DESIGN: For determination of the amounts of BC and LDH released by neutrophils in response to the series of agonists, a total of 30 paired experiments were performed. Twenty experiments were performed to determine the neutrophil BG and LDH responses to the four different treatments of the Pasteurella organisms, and results were compared to the responses of unstimulated control cells. Twelve experiments which compared the ”same day” BG and LDH release responses to agonists and to the 4 treat- ments of Pasteurella organisms, completed the initial studies. In these studies, the levels of agonists that had been defined in studies of oxidative metabolism were used, with the assumption that these levels would also induce sig- nificant enzyme release. For B12BP an initial series of 5 experiments was performed in which the cell requirements and effect of agonist levels were studied (data not shown). These studies confirmed the efficacy of the pre- viously established agonist levels in stimulating specific granule release and the necessity for the use of 2 X 106 129 cells. Subsequently, a series of 10 experiments was con- ducted which compared the "same day” BG, LDH, and B12BP release responses of neutrophils stimulated by the defined series of agonists and by the four treatments of the Pas- teurella organisms. In these latter studies, extreme care was taken to assure that each 2 X 106 cells received equal numbers of OZ and Pasteurella particles per cell. The concentration of soluble agonists, PMA and CI, were the same for all studies. STATISTICAL ANALYSES: Arcsin square root transformation of the percentage release data was performed prior to sta- tistical analyses. Data were analyzed with a computer based statistical program for one-way analysis of variance (ANOVA, OWA)f. Where there were significant effects of treatment in the ANOVA designs at P50.05, least-significant- difference (LSD)f tests were used for comparison of means to determine significance between treatment groups. RESUETS: In the initial comparisons of the responses eli- cited from neutrophils in response to the agonists, calcium ionophore stimulated the highest secretion and/or release of BC and LDH, with these responses differing significantly from those to PMA, OZ, and unstimulated controls (UC) which were not significantly different (N-30). The responses in these assays are shown graphically in Figure 3.1. 1:SAS Institute Inc, Cary, NC. 129A Figure 3.1. Enzyme Release 1: Comparison of the Beta Glucuronidase (BC) and Lactate Dehydrogenase (LDH) Release of Bovine Neutrophils upon Stimulation by Soluble and Particulate Agonists. Analyses by one-way ANOVA (OWA) with comparison of significant means (P50.05) by the least-significant- difference (LSD) test. Each bar represents the means of each enzyme released in response to each agonist with a standard error bar. PMAthorbol ester; CI-cal- cium ionophore; OZ-opsonized zymosan; UC-unstimulated control cells. All percentage release data was transformed by arcsin square root prior to analyses. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. ~ \\\\\\\\B z\\\\\\\\\3 * \\\\\\\\§ -= \- uronldas 131 The series of experiments designated "Enzyme Release 2" compared the release and/or secretion of BC and LDH induced by the four treatments of the Pasteurella organisms and unstimulated control (UC) cells (N-20). These responses showed that there were no significant differences in the quantities of BC released by exposure to live opsonized (PHLO), live (PHL), or dead opsonized (PHDO) when compared to unstimulated control (UC) cells. However, the responses to PHLO, PHL, and PHDO were significantly greater than those to PHD which induced the least BG secretion. PHLO and PHL stimulated the highest release of LDH from neutro- phils and amounts were significantly different from the amounts released in response to PHDO and PHD and by UC (Figure 3.2). Comparisons of the BG and LDH enzyme release responses within the same groups of isolated neutrophils (”same day responses”) stimulated by defined agonists and the Pas- teurella groups (Enzyme Release 3) showed that CI induced the highest (significant) secretion of BC, while the re- sponses to PHLO, PHL, PMA, OZ, PHDO, and of UC were not sig- nificantly different. The BC release in response to PHLO, PHL, and PMA was significantly greater than the release in response to PHD. For LDH, CI stimulated a release that was significantly higher than to PHLO, PHL, and PMA. Only the amounts of enzyme released by exposure to PHLO and PHL dif- fered significantly from the PHDO, UC, PHD, and 02 131A Figure 3.2. Enzyme Release 2: Comparison of the BG and LDH Release of Bovine Neutrophils upon Stimula- tion by Pasteurella haemolytica. Analyses by OWA and LSD. Each bar represents the means of each enzyme released in response to each bacterial preparation with a standard error bar. PHLO-live opsonized, PHLPlive, PHDO-dead opsonized, PHD-dead Easteugeilg haemglztigg organisms; UC-un-stimulated control cells. Abbreviations for agonist treatment groups identical to figure 3.1. All data was arcsin square root transformed prior to analyses. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. 132 Dan mmmcmmotomrma mymflommfl mmmEcoSoEwIm E m: min. more I=._n_ 31m Din. 0: corn. I_In_ Sin. cod mNd Dmd eseaIaJ swfizus sfieIuecued I\ uIs:IJe 133 treatment groups which were not significantly different. (Figure 3.3) The series of 10 experiments designated ”Enzyme Release 4" compared the responses within groups of isolated neutro- phils ("same day responses”) by quantitating the release of BC, LDH, and B12BP induced by incubation with defined con- centrations of agonists and those induced by exposure to the Pasteurella treatment groups. In the BG release as- say, live-opsonized PH stimulated the highest release, dif- fering significantly from the responses to PHL, CI, and PHDO. Neutrophil responses to both treatments also dif- fered significantly from the responses to OZ, PMA, and PHD (Figure 3.4). In the LDH release assay, the quantities of enzyme re— leased after exposure of neutrophils to PHLO, PHL, and CI differed significantly from each other and from the release stimulated by PMA, PHDO, PHD, and OZ. The release induced by PMA, PHDO, and PHD was also significantly greater than that of UC (Figure 3.5). In the quantitation of B12BP secretion and/or release, CI and PHLO induced the highest release with the quantities induced by C1 significantly different from that induced by PMA. The amounts of B12BP released in response to PHLO, while not significantly different from PMA, differed signif- icantly from those induced by OZ and PHL. The responses to PMA and OZ were statistically similar, with only the 133A Figure 3.3. Enzyme Release 3: Comparison of the BG and LDH Release by Bovine Neutrophils in Response to Stimulation by Soluble and Particulate Agonists and by Pasteurella haemolytica. Analyses by OWA and LSD. Each bar represents the means of BC and LDH released in response to agonists and bacterial preparations with a standard error bar. Abbreviations for treatment groups identical to Figures 3.1 and 3.2. All data was arcsin square root transformed prior to analyses. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. .\\\\\\\\\\\\\\\\\\\\\\\\\\\‘ N=IZ - .\\\\\\\\\\\ .\\\\\\\\\\\\\\\\\\‘ .\\\\\\\\\\\\\\\\\\‘ - W \\\\\\\\\\\\\\\\\‘ .\\\\\\\\\\\\\\\\\\V NS 67 :2ng 104/403 890050 :60 67 0625062 894890000 ”6’00? E B-Blucuronidase S Lactate Dehydrogenase I 7 7 c: I: a no II: V; I: I: I: e hzu 6°81U80JBUJ uIsme 134A Figure 3.4. Enzyme Release 4: Comparison of the BG, LDH, and Vitamin BIZ-Binding Protein (B12BP) Release by Bovine Neutrophils in Response to Stimula- tion by Soluble and Particulate Agonists and by Pasteurella haemolytica. Part 1: BG Release. Analyses by OWA and LSD. Each bar represents the means of BC released with a standard error bar. Abbreviations for treatment groups identical to previous figures. All data was arcsin square root transformed prior to analyses. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. ID .I .5 -I%§ —. -\§ ljj mmmmmmm DDDDDDDDD 135A Figure 3.5. Enzyme Release 4: Comparison of the BC, LDH, and B12BP Release by Bovine Neutrophils in Response to Stimulation by Soluble and Particulate Agonists and by Pasteurella haemolytica. Part II: LDH Release. Analyses by OWA and LSD. Each bar represents the means of LDH released with a standard error bar. Abbreviations for treatment groups identical to previous figures. All data was arcsin square root transformed prior to analyses. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significantly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. I-\ .\\\\\\\\\\V0 \\\\\\\\\\\\\\E [._ mmmmm DDDDDDD D ll 2 _I03 \. 2 .J _J 137 responses to PMA differing from the response to PHL. The responses induced by exposure of neutrophils to OZ and PHL were significantly greater than that to PHDO. The response to PHDO was also significantly different from the responses to PHD and of UC which induced no significant release of B12BP (Figure 3.6). DISCUSSION: In the initial studies (enzyme release 1), calcium ionophore induced the highest release of BC but also produced the highest LDH release suggesting that the CI levels used (10"5 M) also induced either membrane dam- age or killed cells (Figure 3.1). It was not possible to conclude that CI induced a selective secretion of primary granules, although the data were suggestive. PMA induced virtually no BC or LDH release, suggesting that at the level of 200 ng the compound had no selective effects on neutrophils to cause the release of primary granules and that it caused little cell toxicity. It has been pre- viously reported that PMA induces a selective release of specific granules in neutrophils.34’53’54 Since the bo- vine neutrophil contains a majority of specific gran- ules99 , major differences in BG release due to an agent which induces selective secretion of specific granules would have been unexpected. OZ stimulation also had no significant effect on primary granule release and was not toxic to cells. 137A Figure 3.6. Enzyme Release 4: Comparison of the BC, LDH, and B12BP Release by Bovine Neutrophils in Response to Stimulation by Soluble and Particulate Agonists and by Pasteurella haemolytica. Part III: B12BP Release. Analyses of arcsin square root transformed data by OWA and LSD. Each bar represents the mean of B12BP released with a standard error bar. Abbrevia- tions for treatment groups identical to previous fig- ures. Horizontal brackets with a centered asterisk (*) enclose treatment groups that differ significant- ly. Horizontal brackets with a subtitle NS enclose groups with no significant differences. N-number of experiments. {a I‘ \\\\\5 IR \\\\\\\\: * \\‘s -""r \\\\\\\\\\\\\\\ LO Z U? PDHD PHD UC N=lD '////A II- _:_. - \\ lllllllll mmmmmmmmm DDDDDDDDD 139 There was extreme variability in the results of BC as- says, because of the extremely low levels of BC in bovine neutrophils. This required the use of four times the num- bers of neutrophils (10 X 106) that are required in BG assays of human neutrophils.54 It was necessary to incu— bate supernatants with substrate for 18 hours to generate quantifiable amounts of reaction product, whereas for other species, 2 to 4 hour incubations have been described. In contrast, the results of assays for LDH release had much less day-to-day variability. The total LDH released by the positive control cells (10 X 106) was always comparable to that of abnormal serum controls. There was excellent reproducibility between experiments (data not shown). In the initial studies (enzyme release 2) of the neutro- phil enzyme release in response to live and dead, opsonized and unopsonized Pasteurella organisms versus the unstim- ulated controls, the responses were incremental with the live opsonized, live, dead opsonized, and dead organisms in- ducing the most to the least responses, respectively, for both BG and LDH (Figure 3.2). These responses were similar to those previously reported in our studies of bovine neu- trophil oxidative metabolism (Chapter 1) and are apparently related to the relative efficiency of neutrophil phagocyto- sis induced by the different treatments of the Pasteur- ella organisms. These data indicate a role for opsonins in the induction of enzyme release and affirms the 140 hypothesis that an undefined factor present in live organ- isms facilitates the uptake of organisms by neutrophils. The increases in LDH release induced by PHLO and PHL were as anticipated, since only viable bacteria are expected to damage or kill neutrophils. In concert with the results for LDH, a portion of the BG released by exposure to PHLO and PHL may have been due to cell lysis. In studies (Enzyme Release 3) which compared the magni- tude of BC and LDH responses caused by the agonists and Pasteurella organisms, CI again induced the highest re- lease of both BG and LDH. There were virtually no (signifi- cant) differences in the quantities of BC released from neu- trophils when induced by the Pasteurella groups, PMA, OZ, or in UC. For LDH, only PHLO and PHL caused signifi- cantly more LDH release than the remaining agonists (Figure 3.3). These data suggest that: 1) exposure of bovine neu- trophils to Pasteurella organisms, even in the presence of opsonins, induces little selective primary granule (BG) release at an approximate cell to bacteria ratio of 5 25:1; 2) a portion of the BG released in response to PHLO and PHL is due to cell damage/lysis as shown by a concomitant eleva- tion in LDH release; and 3) the release of primary granules should have a small or insignificant role in tissue damage attributed to the interaction of Pasteurella organisms with neutrophils. 141 In the studies (enzyme release 4) which compared the re- lease of BC, B12BP, and LDH from neutrophils after stimula- tion by agonists and Pasteurella organisms, the re- sponses were decidedly different and remarkably complex especially for B12BP. In this series of experiments, PHLO and PHL induced the highest release of BC and LDH exceeding the responses to CI. For BC, the quantities of enzyme re- leased in response to PHLO stimulation were significantly greater than those to PHL, CI, and PHDO (Figure 3.4). For LDH, the quantities of enzyme released in response to PHLO, PHL, and CI were all significantly different from each other (Figure 3.5). The parallel responses in both of these assays again suggest that a portion of the BG release was via cell lysis and not entirely due to secretion. As in prior studies opsonization of the Pasteurella organ- isms, live or dead, markedly enhanced the enzyme release by neutrophils. Since the washing of organisms during their preparation should have removed most secreted products, such as endotoxin or cytotoxin we concluded that the LDH re- lease was due to cell death and cell lysis induced by the ingested pathogens. Calcium ionophore and PHLO stimulated the greatest re- lease of specific granules from neutrophils, with the re- sponses being nearly identical. The responses to CI were significantly higher than for PMA (Figure 4-6). This re- d54,55 sponse was unexpected since PMA has been reporte to 142 induce a selective release of specific granules. Only the report by Gallin indicated similar responses to CI.34 The propensity of CI to induce membrane damage (increased LDH release) may have also increased the quantities of BlZBP released. A similar conclusion was reached for the neutrophil responses to live organisms (PHL) with the ef- fect enhanced by opsonization (PHLO). Therefore, it was not possible to differentiate between the amounts of B12BP released due to neutrophil lysis and those due to secre- tion. In contrast to the previous data the release of B12BP stimulated by PMA appeared to be principally secre- tory, at the concentrations used (200 ng), since there was comparatively little LDH released. In initial studies of enzyme release, the concentra- tions (particle/cell ratios) of particulate agonists used were the same as in the studies of neutrophil oxidative metabolism, and these concentrations were maintained, al- though cell numbers increased from 2 X 106 to 10 X 106. In the comparative studies of the three enzyme assays, extreme care was taken to assure that equal quanti- ties of particulate agonists were provided for each 2 X 106 cells. Differences in the neutrophil responses in- duced by particulate agonists, e.g. Pasteurella, in initial and comparative response experiments may be partial- ly explained by the presence of fewer particles per cell in the initial studies. 143 The differences in the responses to soluble agonists between initial and comparative protocols are less easily explained. In initial studies of oxidative metabolism, we found that by using the same concentrations of soluble agon- ists (PMA, CI) and doubling cell numbers there were vir- tually no differences in cell oxidative responses (data not shown). We postulated that the number of molecules of stim- ulant (agonist) per cell was not sufficient to induce maxi- mal oxidative responses of all cells. In neutrophils and other cells, it is well known that the affinity states of receptors, and perhaps enzymes, may modulate cell re- sponses, including enzyme release. Regardless of the mechanisms of release, lysis or secre- tion, the quantities of specific granule components re- leased in response to live opsonized, live, or dead opson- ized Pasteurella organisms were substantial. Since the responses to OZ were also comparatively high, the presence of opsonized particles, live or dead, pathogenic or non- pathogenic, may facilitate the release of specific granules by bovine neutrophils. The high levels of specific granule enzymes in bovine neutrophils combined with a high rate of release or secretion in response to Pasteurella organ- isms could have a major role in the characteristic pulmon- ary lesions of pneumonic pasteurellosis. Neutrophil specific granules have been characterized by Gallin34, as "the fuse that ignites the inflammatory 144 process." Although primary granules are also released by bovine neutrophils in response to the Pasteurella or- ganism, partially due to cell lysis, relatively low numbers (compared to specific granules) and relatively low release in response to the presence of the organisms seem to pre- clude significance in pneumonic pasteurellosis. Several investigators have implicated the release of primary gran- ules with the modulation or "turn off" of the inflammatory response34’35 and have disclaimed the role of primary granule components in tissue injury. The myeloperoxidase- halide pathway has been shown to facilitate the disruption of the NAD(P)H-oxidase enzyme complex leading to the cessa- tion of superoxide production.35 Primary granule pro- ducts have also been shown to destroy alpha-l-antitrypsin, prostanoid metabolites, and C5a.185’186 CONCLUSIONS: In summary, bovine neutrophils respond to Pasteurella haemolytica by the secretion and/or release of primary and specific granules, a process enhanced by op- sonins. A portion of the enzyme release of both granule subtypes is caused by the killing or lysis of cells induced by the ingested Pasteurella organisms. Because speci- fic granules are present in greater numbers in bovine neu- trophils99 and have a greater propensity for secretion and/or release, regardless of mechanism, in response to the presence of Pasteurella organisms, we suggest that the 145 specific granules of bovine neutrophils have a major role in the pathogenesis of pneumonic pasteurellosis. Future studies should involve the determination of the role of in vitro and in vivo blockers of the secre- tion of specific granules as a means of ameliorating pulmon- ary injury in pneumonic pasteurellosis. Since prior stud- ies have shown that endotoxin facilitates the oxidative re- sponses of bovine neutrophils, the effects of endotoxin on the release of neutrophil specific granules in combination with exposure to Pasteurella organisms should also be explored. SUMMARY My research into the influence of Pasteurella haemolytica and its products on the function of bovine neutrophils defined the in vitro oxidative responses of isolated bovine neutrophils by means of luminol-enhanced chemiluminescence (LECL) and the production of superoxide anion when stimulated by selected agonists and agonist con- centrations. These responses were characterized by marked day-to-day variability within and between individual ani- mals and were partially dependent upon the.age of the exper- imental animals, and upon the mechanisms by which individ- ual agonists activate neutrophils. Live Pasteurella organisms and/or opsonins enhanced the LECL and superoxide responses of stimulated neutrophils, although in compara- tive studies these responses were exceeded by non-biologi- cal entities. In my endotoxin experiments, I confirmed that the presence of trace quantities of Pasteurella haemolytica (PH) or Escherichia coli (EC) endotoxins stimulated bovine neutrophils to produce increased quantities of superoxide anion in response to phorbol ester (PMA) and calcium ionophore (CI). These results imply that endotoxin exposure selectively enhances protein kinase C and Ca2+- mediated pathways of neutrophil transduction, whereas there 146 147 were no differences the the pathways mediated by particu- lates [latex or opsonized zymosan (02)]. Diametrically, endotoxin exposure markedly decreased the LECL responses of neutrophils to PMA and CI, but again had no effect on the responses to particulate agonists. My experiments also determined that there was no species-specific effect of PH endotoxin. My experiments defining the magnitude of B-glucuroni- dase from primary granules, vitamin BIZ-binding protein from specific granules, and cytosolic lactate dehydrogenase secretion/release by stimulated bovine neutrophils repre- sent, to my knowledge, the first studies of this kind in the bovine species. These experiments confirmed that the near selective release of specific granules in response to the Pasteurella organism was enhanced by the presence of live bacteria and/or opsonins. They confirmed that the organisms are capable of injuring neutrophils after they are phagocytosed, as indicated by the concomitant enhanced release of cytosolic LDH. In summary, Pasteurella haemolytica induces bovine neutrophils to produce reactive oxygen species including superoxide anion, perhaps enhanced by low levels of its endotoxin, and facilitates a marked release of specific granules, processes enhanced by opsonic factors. All of these factors may have possible roles in the disease. 10. ll. 12. LIST OF'REFERENCES PART I: MECHANISMS OF NEUTROPHIL.ACTIVATION (TRANSDUCTION) . Klebanoff SJ. Clark RA. W11: W Qlinica1_21§9rderg. North Holland Publishing Company, Amsterdam, New York, 1978. . Weissmah G. editor. WW. Elsevier/North Holland Biomedical Press, Amsterdam, New York, 1980. - Movat HZ editor. WW- ssnsitixisx: Whamm- 2nd ed. Harper and Row, Hagerstown, 1979. . Boxer GJ, Curnette JT, Boxer LA. Polymorphonuclear leukocyte function. Hosp Pract 1985;34:69-90. . Fantone JC, Ward PA. Role of oxygen-derived free radi- cals and metabolites in leukocyte-dependent inflamma- tory reactions. Am J Pathol 1982;107:398-417. . Lynn WS. Activation of human inflammatory cells. In: Lynn WS. Ed. 1W. CRC Press, Boca Raton, Florida 1983;85-101. . Berridge MJ. Inositol trisphosphate and diacylglycerol as second messengers. Biochem J 1984;220:345-360. . Marx JL. Polyphosphoinositide research updated. Science 1987; 974-976. . Berridge ME, Irvine RF. Inositol trisphosphate, a novel second messenger in cellular signal transduc- tion. Nature 1984; 312:315-321. Ryu SH, Lee SY, Lee K. Catalytic propertigs of inosi- tol triphosphate kinase: Activation by Ca + and calmodulin. FASEB J l987;1:388-393. Becker EL, Kermode JC, Naccache PH, et. a1. Pertussis toxin as a probe of neutrophil activation. Federation Proc 1986;45:2151-2155. Fletcher MP, Gallin JI. Degranulating stimuli increase the availability of receptors on human neutrophils for the chemoattractant fmet-leu-phe. J Immunol 1980; 124:1585-1588. 148 l3. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 149 Fletcher MP, Seligmann BE, Gallin JI. Correlation of human neutrophil secretion, chemoattractant receptor mobilization and enhanced functional activity. J Immunol 1982; 128:941-948. Fletcher MP, Gallin JI. Human neutrophils contain an intracellular pool of putative receptors for the chemoattractant N-formyl-methionyl-leucyl-phenylala- nine. Blood 1983; 62:792-799. Hoffstein ST, Friedman RS, Weissman G. Degranulation, membrane addition and shape change during chemotactic factor-induced aggregation of human neutrophils. J Cell Biol 1982; 95:234-241. Naccache PH, Molski MM, Volpi M. Unique inhibitory profile of platelet activating factor induced calcium mobilization and polyphosphoinositide turnover and granule enzyme secretion in rabbit neutrophils toward pertussis toxin and phorbol ester. Biochem Biophy Res Commun 1985;130: 677-684. Gomperts BD, Barrowman MM, Cockcroft S. Dual role for guanine nucleotides in stimulus-secretion coupling. Federation Proc 1986;45: 2156-2161. Nishizuka Y. Studies and Perspectives of Protein Kinase C. Science 1986;233:305-312. Nishizuka Y. The role of protein kinase C in cell surface signal transduction and tumour promotion. Nature 1984;308:693-698. Tauber AI. Protein kinase C and the Activation of the Human neutrophil NADPH-Oxidase. Blood 1987;69:711-720. Garcia-Gil M, Alonso F, Alvarez-Chiva V, et.al. Phos- pholipid turover during phagocytosis in human polymor- phonuclear leukocytes. Biochem J 1982; 206:67-72. Volpi M, Naccache PH, Molski TFP, et. a1. Pertussis toxin inhibits f-Met-Leu-Phe- but not phorbol ester- stimulated changes in rabbit neutrophils: Role of G proteins in excitation response coupling. Proc Natl Acad Sci USA 1985;82:2708-2715. Cox CC, Dougherty RW, Ganong BR, et.al. Differential stimulation of the respiratory burst and lysosomal enzyme secretion in human polymorphonuclear leukocytes by synthetic diacylglycerols. J Immunol 1986; 136:4611-4617. 150 24. Liles WC, Meier KE, Henderson WR. Phorbol myristate 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. acetate and the calcium ionophore A23187 synergistic- ally induce release of LTB4 by human neutrophils: Involvement of protein kinase C in regulation of the 5-lipoxygenase pathway. J Immunol 1987;138:3396-3402. Melloni E, Pontrmoli S, Michetti M, et. al. Binding of protein kingse C to neutrophil membranes in the pre sence of Ca + and its activation by a Ca +-requir- ing proteinase. Proc Natl Acad Sci USA 1985; 82:6435-6442. Pontremoli S, Melloni E, Michetti M, et.al. The involvement of calpain in the activation of protein kinase C in neutrophils stimulated by phorbol myristic acid. J Biol Chem 1986;261:4101-4107. Pontremoli S, Melloni E, Michetti M, et. a1. Biochemi cal responses in activated haman neutrophils mediated by protein kinase C and a Ca requiring proteinase. J Biol Chem 1986;261:8309-8313. Tauber AI, Borregard N, Simons E, et. a1. Chronic granulomatous disease: A syndrome of phagocyte oxidase deficiencies. Medicine 1983;62:286-293. Tauber AI. Current views of neutrophil dysfunction: An integrated clinical perspective. Am J Med 1981; 70:1237-1247. Green TR, Wu DE. The NADPH:02 Oxidoreductase of Human neutrophils. J Biol Chem 1986; 261:6010-6015. Gennaro R, Florio C, Romeo D. Co-activation of Protein kinase C and NADPH oxidase in the plasma membrane of neutrophil cytoplasts. Biochem Biophys Res Commun 1986;134:305-312. Ohno Y, Seligmann BE, Gallin JI. Cytochrome b transla- cation to the human neutrophil plasma membranes and superoxide release: Differential effects of N-formyl- methionylleucylphenylalanine, phorbol myristate ace tate and A23187. J Biol Chem 1985; 260:2409-2414. Cox JA, Jeng AY, Blumberg PM, et. a1. Comparison of subcellular activation of the human neutrophil NADPH-oxidase by arachidonic acid, sodium dodecyl sulfate (SDS) and phorbol myristate acetate (PMA). J Immunol 1987;138:1884-1888. Gallin JI. Neutrophil Specific Granules: A Fuse That Ignites The Inflammatory Response. Clin Res 1984;32:320-328. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 151 Edwards SW, Swan TF. Regulation of superoxide genera- tion by myeloperoxidase during the respiratory burst of human neutrophils. Biochem J 1986; 237:601-604. Borregard N, Schwartz JH, Tauber AI. Proton secretion by stimulated neutrophils: Significance of hexose monophosphate shunt activity as source of electrons and protons for the respiratory burst. J Clin Invest 1984; 77:782-788. Tauber AI, Babior BM. Evidence for hydroxyl radical production by human neutrophils. J Clin Invest 1977; 60:374-379. Weiss SS, Rustagi PK, LoBuglio AF. Human granulocyte generation of hydroxyl radical. J Exp Med 1978; 147:316-324. Allen RC, Stjernholm RL, Steele RH. Evidence for the generation of an electronic excitation state(s) in human polymorphonuclear leukocytes and its participa- tion in bacteriacidal activity. Biochem Biophys Res Commun 1972; 47:679-684. Cheson BD, Christensen RL, Sperling R, et.a1. The origin of chemiluminescence of phagocytosing granulo- cytes. J Clin Invest 1976; 58:789-796. Anderson BR, Brendzel AM, Lint TF. Chemiluminescence spectra of human myeloperoxidase and polymorphonuclear leukocytes. Infect Immun 1977; 17:62-66. Rosen H, Klebanoff SJ. Formation of singlet oxygen by the myeloperoxidase-mediated antimicrobial system. J Biol Chem 1977; 252:4803-4810. Klebanoff SJ. Myeloperoxidase-halide-hydrogen peroxide antibacterial system. J Bacteriol 1968; 95:2131-2138. Harrison JE, Schultz J. Studies on the chlorinating activity of myeloperoxidase. J Biol Chem 1976; 251:1371-1374. Fee JA, Valentine JS. Chemical and physical properties of superoxide. In: Superoxide and Superoxide dismu- tases. Michelson AM, McCord JM, Fridovich I (eds), Academic Press, New York, 1977, pp 19-60. Nilsson R, Pick FM, Bray RC. ESR studies on reduction of oxygen to superoxide by some biochemical systems. Biochem Biophy Acta 1969; 192:145-148. 47. 48. 49. 50. 51. 52. 53. 54. 55. 56. 57. 58. 152 Ambruso DR, Johnston RB. Lactoferrin enhances hydroxyl radical production by human neutrophil particulate fractions, and an enzymatic generating system. J Clin Invest 1981; 67:352-360. Babior BM. Oxygen-dependent microbial killing by phagocytes. N Engl J Med 1978; 298:659-668, 721-725. Piatt JF, O'Brien PJ. Singlet oxygen formation by a peroxidase, H20 , and halide system. Eur J Biochem 1979; 9 :323-332. Maridonneau-Parini I, Tringale SM, Tauber AI. Identi- fication of distinct activation pathways of the human neutrophil NADPH-oxidase. J Immunol 1986; 137:2925-2931. DeVirgilio F, Lew DP, Pozzan T. Protein kinase C acti- vation of physiological processes in human neutrophils at vanishingly small cytosolic Ca++ levels. Nature 1984; 310:691-695. Maridonneau-Parini I, Tauber AI. Activation of NADPH-oxidase by arachidonic acid involves phospholi- pase A2 in intact human neutrophils, but not in a cell free system. Biochem Biophs Res Commun 1986; 138:1099-1105. Babior BM, Cohen HJ. Measurement of neutrophil func- tion: Phagocytosis, degranulation, the respiratory burst and bacterial killing. In: Lgukggy§g_fiunctign. Cline MJ (ed) Churchill Livingstone, New York 1981, pp 1-39. Metcalf JA, Gallin JI, et a1. Lahg;g§gzy_ngngal_gf Neg;;gphil_fignggign‘ Raven Press, New York. 1986. Thrush MA, Wilson ME, Van Dyke K. The generation of chemiluminescence (CL) by phagocytic cells. Methods in Enzymol 1978;57:462. Nelson RD et a1. Chemiluminescence response of human leukocytes: Influence of medium components on light production. Infect Immun 1977;17:513-520. Westrick MA, Shirley PS, De Chatelet LR. Generation of chemiluminescence by human neutrophils exposed to soluble stimuli of oxidative metabolism. Infect Immun 1980;30:385-390. Cheung K, Archibald AC, Robinson MF. Luminol-dependent chemiluminescence produced by neutrophils stimulated by immune complexes. Aust J Exp Biol Med Sci 1984; 62(Pt 4):403-419. 59. 60. 61. 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 153 Allen RC, Loose LD. Phagocytic activation of a luminol-dependent chemiluminescence in rabbit alveolar and peritoneal macrophages. Biochem Biophys Res Commun 1976;69:245-252. Wilson ME, Trush MA, VanDyke K, et.al. Luminol-depen- dent chemiluminescence of opsonophagocytic dysfunc- tions. J Immunol Meth 1978; 23:315-326. Dormandy TL. Free-radical oxidation and antioxidants. Lancet 1978; 1:647-650; Al Timini DJ, Dormandy TL. Inhibition of lipid autoox- idation by human ceruloplasmin. Biochem J 1977; 168:283-288. Plonka A, Metodiewa D. ESR evidence of superoxide radical dismutation by human ceruloplasmin. Biochem Biophys Res Commun 1979; 95:978-984. Stocks J, Gutteridge JM, Sharp RJ, et.al. The inhibi- tion of lipid autooxidation by human serum and its relation to serum proteins and alpha-tocopherol. Clin Sci Mol Med 1974; 47:223-232. Chow CK. Distribution of tocopherol in human plasma and red blood cells. Am J Clin Nut 1975; 28:756. Machlin LJ (ed). WW. Marcel Dekker, Inc., 1980. Michelson AM, McCord JM, Fridovich I (eds) fiupgrgxigg gnd Superoxide Qismutasgs. Academic Press, New York, 1977. Fridovich 1. Oxygen radicals, hydrogen peroxide and oxygen toxicity. In: Free Radicals in biology. Vol I. Pryor WA (ed). Academic Press, New York, 1976, pp 239-277. Lawrence RA, Burk RE. Glutathione peroxidase activity in selenium-deficient rat liver. Biochem Biophys Res Commun 1978; 71:952-958. Lawrence RA, Burk RE. Species, tissue, and subcellular distribution of non-selenium dependent glutathione peroxidase activity. J Nutr 1978; 108:211-215. Bainton DF, Farquhar MG. Differences in enzyme content of azurophil and specific granules of polymorphonuc- lear leukocytes. II. Cytochemistry and electron micro- scopy of bone marrow cells. J Cell Biol 1968; 39:299-317. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. 82. 154 Stossel TP, Hartwig JF, Yin HL, et. al. The Motor of Leukocytes. Federation Proc 1984;43:2760-2763. Morrison DC, Ulevitch RJ. The effects of bacterial endotoxins on host mediation systems. Am J Pathol 1978; 93:527-617; Haslett C, Worthen GS, Giclas PC, et. al. The Pulmo- nary Vascular Sequestration of Neutrophils in Endo- toxemia is initiated by an effect of endotoxin on the neutrophil in the rabbit. Am Rev Respir Dis 1987;136:9-18. Hoffstein ST, Gennaro DE, Manzi RM. Surface contact inhibits neutrophil superoxide generation induced by soluble stimuli. Lab Invest 1985; 52:515-522. Haslett CH, Guthrie LA, Kopaniak MM, et.al. Modulation of multiple neutrophil functions by preparative methods or trace concentrations of bacteria lipopoly- saccharide. Am J Pathol 1985; 119:101-110. Guthrie LA, McPhail LC, Henson PH, et. a1. Priming of Neutrophils For Enhanced Release of Oxygen Metabolites By Bacterial Lipopolysaccharide: Evidence for Increas- ed Activity of the Superoxide-producing Enzyme. J Exp Med 1984;160:1656-1671. Graham RC, Jr., Karnovsky MJ, Shafer AW, et.a1. Metabolic and morphological observations on the effect of surface-active agents on leukocytes. J Cell Biol 1967; 32:629-633. Proctor RA. Endotoxin in 11;;9 interactions with human neutrophils: depression of chemiluminescence, oxygen consumption, superoxide production, and killing. Infect Immun 25:912-918. Wilson ME, Jones DP, Munkenbeck P, et. a1. Serum-de- pendent and -independent effecs of bacterial lipopoly- saccharides on human neutrophil oxidative capacity in vitro. J Reticuloendoth Soc 1982; 31:43-48. Spirer Z, Zakuth A, Golander N, et.al. The effect of tuftsin on the nitrous blue tetrazolium reduction of normal human polymorphonuclear leukocytes. J Clin Invest 1975; 55:198-203. Dahinden C, Fehr AJ. Granulocyte activation by endo- toxin. 11. Role of granulocyte adherence, aggregation, and effect of cytochalasin B, and comparison with formylated chemotactic peptide-induced stimulation. J Immunol 1983; 130:863-869. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 155 Dahinden CA, Fehr AJ, Hugli TE. Role of cell surface contact in the kinetics of superoxide production by granulocytes. J Clin Invest 1983; 72:113-119. Dahinden C, Galanos C, Fehr J. Granulocyte Activation By Endotoxin I. Correlation between Adherence and Other Granulocyte Functions, and Role of Endotoxin Structure on Biologic Activity. J Immunol 1983; 130:857-862. Wilson ME. Effect of Bacterial Endotoxins on Neutro- phil Function. Rev Infect Dis l985;7:404-4l8. Morrison DC. Bacterial endotoxins and pathogenesis. Rev Infect Dis 1983; 5:(suppl 4):S733-747. Nies AS, Greineder DR, Cline MJ, et.al. The divergent effects of endotoxin on human plasma and leukocytes. Biochem Pharmacol 1971; 20:39-46. Mesrobeanu I, Bona C, Mesrobeanu L. Pinocytosis by leukocytes of ”O” endotoxins. Exp Cell Res 1964; 36:434-438. Fehr J, Jacob HS. In xigrg granulocyte adherence and in yivg margination: two associated complement depen- dent functions. Studies based on the acute neutro- penia of filtration leukophoresis. J Exp Med 1977; 146:641-652 Fehr J, Dahinden C. Modulating influence of chemotac- tic factor-induced cell adhesiveness on granulocyte function. J Clin Invest 1979; 64:8-16. Oseas R, Yang H-H, Baener RL, et.al. Lactoferrin: a promoter of polymorphonuclear leukocyte adherence. Blood 1981; 57:939-945. Modrzakowski MC, Spitznagel JK. Bacteriacidal activity of fractionated granule contents from human polymor- phonuclear leukocutes: antagonism of granule cationic proteins by lipopolysaccharide. Infect Immun 1979; 25:597-602. Guthrie LA, Johnston RB Jr. Neutrophil "priming" by bacterial endotoxin (LPS)[abstract]. Fed Proc 1982; 41:933. 94. 95. 96. 97. 98. 99. 100. 101. 102. 103. 104. 105. LIST OF REFERENCES PART II: THE BOVINE NEUTROPHIL Carlson GP, Kaneko JJ. Intravascular granulocyte kine- tics in developing calves. Am J Vet Res 1975; 36:421-425. Schalm OW. Veteringr1_flgmagglggy. 2nd Ed. Lea and Febiger, Philadelphia, PA. 1965. Greatorex JC. Studies on the haematology of calves from birth to one year of age. Br Vet J 1957; 113:29-65. Holman HH. Changes associated with age in the blood picture of calves and heifers. Br Vet J 1956; 112:91-104. Rausch PG, Moore TG. Granule enzymes of polymorpho- nuclear neutrophils: A phylogenetic comparison. Blood 1975;46:913-919. Gennaro RC, Scheider C, De Nicola G, et a1. Biochemi- cal properties of bovine granulocytes. Proc Soc Exp Biol Med 1978;157:342-347. Gennaro RC, Dewald B, Horisberger U, et.a1. A novel type of cytoplasmic granule in bovine neutrophils. J Cell Biol 1983;96:1651-1661. Baggiolini M, Horisberger U, Gennaro R et al. Ident- ification of three types of granules in neutrophils of ruminants: Ultrastructure of circulating and maturing cells. Lab Invest 1985;52:151-158. Roth JA, Kaberle ML. Evaluation of bovine polymorpho- nuclear leukocyte function. Vet Immunol Immunopathol l981;2:157-l74. Carlson GP, Kaneko JJ. Isolation of leukocytes from peripheral blood. Proc Soc Exp Biol Med 1973; 142:853-856. Riding GA, Willadsen R. Simultaneous isolation of bovine eosinophils and neutrophils on gradients of percoll. J Immunol Meth 1981;46:113-119. Roth JA, Kaberle ML. Isolation of neutrophils and eosinophils from the peripheral blood of cattle and comparison of their functional activities. J Immunol Meth 1981;45:153-164. 156 106. 107. 108. 109. 110. 111. 112. 113. 114. 115. 116. 157 Chambers WH, Taylor JH, Klesius PH. Isolation of bovine polymorphonuclear leukocytes by density gradient centrifugation. Vet Immunol Immunopathol 1983;5zl97-202. Roth JA, Kaeberle ML, Griffith RW. Effects of bovine viral diarrhea virus infection on bovine polymorpho- nuclear leukocyte function. Am J Vet Res 1981; 42:244-250. Roth JA, Kaeberle ML. Suppression of neutrophil and lymphocyte function induced by a vaccinal strain of bovine viral diarrhea virus with and without the administration of ACTH. Am J Vet Res 1983; 44:2366-2372. Roth JA, Bolin SR, Frank DE. Lymphocyte blastogenesis and neutrophil function in cattle persistently infect- ed with bovine viral diarrhea virus. Am J Vet Res 1986; 47:1139-1141. Roth JA, Kaeberle ML. Effects of ACTH administation on bovine polymorphonuclear leukocyte function and lymphocyte blastogenesis. Am J Vet Res 1982; 43:412-416. Roth JA, Kaeberle ML. Effect of in vivo dexamethasone administration on in vitro bovine polymorphonuclear leukocyte function. Infect Immun 1981; 35:434-441. Roth JA, Xaeberle ML. In vivo effect of ascorbic acid on neutrophil function in healthy and dexamethasone- treated cattle. Am J Vet Res 1985; 2434-2436. Roth JA, Kaeberle ML, Hsu WH. Effect of estradiol and progesterone on lymphocyte and neutrophil functions in steers. Infect Immun 1982; 35:997-1002. Ryu H, Kaeberle ML, Roth JA, et.a1. Effect of type A Pasteurella mglgggigg fractions on bovine polymorpho- nuclear leukocyte functions. Infect Immun 1984; 43:66-71. Hubbard RD, Kaeberle ML, Roth JA, et.a1. flggmgphilgg figmngg-induced interference with bovine neutrophil functions. Vet Microbiol 1986; 12:77-85. Chiang YW, Kaeberle ML, Roth JA. Identification of suppressive components in "Haemophilus somnus" fractions which inhibit bovine polymorphonuclear leukocyte function. Infect Immun 1986; 52:792-797. 117. 118. 119. 120. 121. 122. 123. 124. 125. 126. 127. 158 Czuprynski CJ, Hamilton HL. Bovine neutrophils ingest but do not kill flggmgphilgg §gmng§. Infec Immun 1985; 50:431-436. Bertram TA, Canning PC, Roth JA. Preferential inhibi- tion of primary granule release from bovine neutrophil by a 51392113 gggrgug extract. Infect Immun 1986; 52:285-292. Canning PC, Roth JA, Tabatabai LB, et.al. Isolation of components of firuggllg ghgxggs responsible for inhibi tion of function in bovine neutrophils. J Infect Dis 1985; 152:913-921. Canning PC, Roth JA, Deyoe BL. Release of 5'-guano- sine monophophate and adenine by Brucella abortus and their role in the intracellular survival of the bacteria. J Infect Dis 1986; 154:464-470. Steinbeck MJ, Roth JA, Kaeberle ML. Activation of bovine neutrophils by recombitant gamma interferon. Cell Immunol 1986; 1:137-144. Lukacs K, Roth JA, Kaeberle ML. Activation of neutro- phils by antigen-induced lymphokines with emphasis on antibody-independent cytotoxicity. J Leukocyte Biol 1985; 38:557-572. Ohmann HB, Babiuk LA. Effect of bovine recombitant alpha-1 interferon on inflammatory responses of bovine phagocytes. J Interferon Res 1984; 4:249-263. Ohmann HB, Babiuk LA. Alteration of some leukocyte functions following in vivg and in zitrg exposure to recombitant alpha- and gamma-interferon. J Interferon Res 1986; 6:123-136. Roth JA, Kaeberle ML, Hubbard RD. Attempts to use thiabendazole to improve the immune response in dexa- methasone-treated or stressed cattle. Immunopharmacol 1984; 8:121-128. Kaeberle ML, Roth JA. Effects of thiabendazole on dexamethasone-induced suppression of lymphocyte and neutrophil function in cattle. Immunopharmacol 1984; 8:129-136. Hauser MA, Koob MA, Roth JA. Variation of neutrophil function with age in calves. Am J Vet Res 1986; 47:152-153. 128. 129. 130. 131. 132. 133. 134. 135. 136. 137. 159 Slocombe RF, Malark J, Ingersoll R, et a1. Neutrophil depletion of calves with hydroxyurea: methods and clinical and pathologic effects. Am J Vet Res 1986; 47:2313-2317. Slocombe RF, Malark J, Ingersoll R, et a1. Acute pulmonary injury in calves caused by Egggggrglla hggmglytlgg requires neutrophils. Am J Vet Res 1985; 46:2253-2258. Berggren RA, Baluyut CS, Simonson RR, et.a1. Cytotoxic effects of Eggggurgllg hagmglygigg on bovine neutro- phils. Am J Vet Res 1981; 42:1383-1388. Czuprynski CJ, Hamilton HL, Noel EJ. Ingestion and killing of Pasteurella haemelxtisa by bovine neut- rophils in vitro. Vet Microbiol 1987; 14:61-74. O'Brien JK. Duffus WPH. Pasteurella hasmellsisa cytotoxin: relative susceptibilities of bovine leucocytes. Vet Microbiol 1987; 13:321-334. Chang YF, Renshaw HW, Martens RJ, et.al. Bagtgurella hggmglygigg leukotoxin: chemiluminescent responses of peripheral blood leukocytes from several different mammalian species to leukotoxin- and opsonin-treated living and killed Pasteurella haemelxsisa and Staphylggogcus augggg. Am J Vet Res 1986; 47:67-74. Chang YF, Renshaw HW, Augustine JL. Bovine pneumonic pasteurellosis: chemiluminescent response of bovine peripheral blood leukocytes to living and killed Eastssrella haemglxsisa. Pasteurella multasiga. and Escherichia 9211. Am J Vet Res 1985; 46:2266-2271. Chang YF, Renshaw HW. teu hggmglygigg leuko- toxin: Comparison of 5 -chromium-release, trypan blue dye exclusion, and luminol-dependent chemilumi- nescence-inhibition assays for sensitivity in detecting leukotoxin activity. Am J Vet Res 1986; 47:134-138. Forsell JH, Kateley JR. Subchronic administration of technical pentachlorophenol in lactating dairy cattle: immunotoxicologic evaluation. J Toxicol Environ Health 1981; 8:543-558. Phillips TR, Yang WC, Schultz RD. The effects of glucocorticoids on the chemiluminescence response of bovine phagocytic cells. Vet Immunol Immunopathol 1987; 14:245-256. 138. 139. 140. 141. 142. 143. 144. 145. 146. 147. 148. 149. 160 Phillips TR, Yang WC, Schultz RD. In XLEIQ effects of prostaglandin E1, prostaglandin E2, indomethacin, histamine, and tuftsin on chemiluminescence response of bovine polymorphonuclear leukocytes. Vet Immunol Immunopathol 1987; 14:233-244. Korhonen HJ, Reiter B. Production of H20 by bovine blood and milk polymorphonuclear eucocytes. Acta Microbiol Pol 1983; 32:53-64. Young S, Beswick P. A comparison of the oxidative reactions of neutrophils from a variety of species when stimulated by opsonized zymosan and F-met-leu-phe. J Comp Pathol 1986; 96:189-196. McGuire TC, Musoke AJ. Biologic activities of bovine IgG subclasses. Adv Exp Med Biol 1981;137:359-366. McGuire TC, Musoke AJ, Kurtti T. Functional properties of bovine IgGl and IgGZ: interaction with complement, macrophages, neutrophils and skin. Immunology 1979; 38:249-256. Howard CJ, Taylor G. Interactions of mycoplasmas and phagocytes. Yale J Biol Med 1983;56:643-648. Williams RM, Hill AW. A role for IgM in the in zitrg opsonisation of figgphylgggggug guregs and Egghggighig £91; by bovine polymorphonuclear leucocytes. Res Vet Sci 1982;33:47-53. Mueller R, Boothby JT, et a1. Change of complement values in calves during the first month of life. Am J Vet Res 1983;44:747-750. Renshaw HW, Eckblad WP, et al. Levels of total hemoly- tic complement activities in paired dairy cow-newborn calf sera. Immunology 1978;34:801-805. Renshaw HW, Everson DO. Classical and alternate com- plement pathway activities in paired dairy cow-newborn calf sera. Comp Immunol Microbiol Infect Dis 1980; 1:259-267. Rice CE, Duhamel L. A comparison of the complement, conglutinin and natural anti-sheep red cell antibody titers of newborn and older calves. Can J Comp Med 1957; 21:109-116. Lombardo PS, Todhunter DA, et al. Effect of colostrum ingestion on indices of neutrophil phagocytosis and metabolism in newborn calves. Am J Vet Res 1979; 40:362-368. 150. 151. 152. 153. 154. 155. 156. 157. 158. 159. 160. 161. 161 Woodard LF, Eckblad WP, et al. Effects of maternal protein-energy malnutrition and cold stress on neutrophil function of bovine neonates. Am J Vet Res 1980;41:1208-1211. Reffett JK, Spears JW, Brown TT, Jr. Primary and secondary immune responses of selenium deficient calves challenged with IBR virus. Fed Proc [abstr] 1987;46:1154. Gyang E0, Stevens JB, et al. Effects of selenium-vita- min E injection on bovine polymorphonucleated leuko- cytes phagocytosis and killing of figaphylggggggg agreug. Am J Vet Res 1984;45:175-177. Arthur JR, Boyne R, et al. The production of oxygen- derived radicals by neutrophils form selenium defi- cient cattle. FEBS Lett 1981;135:187-190. Boyne R, Arthur JR. Alteration of neutrophil function in selenium-deficient cattle. J Comp Pathol 1979. 89:151-158. Boyne R, Arthur JR. Effects of selenium and copper deficiency on neutrophil function in cattle. J Comp Pathol 1981;91:271-276. Boyne R, Arthur JR. Defective leucocyte function in selenium-deficient cattle. Proc Nutr Soc 1979;38:14A. Williams MR, Hibbitt KG, et a1. Further studies on the variation among cows, bulls, and calves in the ability of their blood polymorphonuclear leukocytes to kill ,§§§phylgggggg§ gurggs. Br Vet J 1984; 140:307-313. Anderson KL, Smith AR, Shanks RD, et.al. Endotoxin-in- duced bovine mastitis: immunoglobulins, phagocytosis, and effect of flunixin meglemine. Am J Vet Res 1986; 47:2405-2410. Confer AW, Simons KR. Effects of Eaggguggllg hggmglx- giga lipopolysaccharide on selected functions of bovine leukocytes. Am J Vet Res 1986;47:154-157. Lewis RA, Austin KF. The biologically active leuko- trienes. Biosynthesis, metabolism, receptors, func- tions, and pharmacology. J Clin Invest 1984; 73:889-897. O'Flaherty JT. Lipid mediators of acute allergic and inflammatory reactions.In: Lynn WS, ed. Inflammatgzy Cells gng Lung Disease, CRC Press, Boca Raton, Florida, 1983;1-29. 162. 163. 164. 165. 166. 167. 168. 169. 170. 171. 172. 173. 174. 162 O'Flaherty JT. Lipid mediators of inflammation and allergy. Lab Invest 1982;47:314-329. Ford-Hutchison AW. Leukotrienes: Their formation and role as inflammatory mediators. Fed Proc 1985; 44:25-29. Tizard I. Veterinary_lmmunglggy. 2nd ed. 1982 W.B. Saunders Co. Eiermann GJ, Dickey BF, Thrall RS. Polymorphonuclear leukocyte participation in acute oleic acid-induced lung injury. Am Rev Resp Dis 1983;128:845-850. Daugherty HW, Hen A. The role of polymorphonuclear peroxidase-dependent oxidants in inflammation. Agents-Actions Suppl l980;7:167-173. Schraufstatter IU, Revak SD, Cachrane CG. Proteases and oxidants in experimental pulmonary inflammatory injury. J Clin Invest 1984;73: 1175-1184. Ward PA, Sulavik BA, Johnson KJ. Activated rat neut- rophils: correlation of arachidonate products with enzyme secretion but not with 02- generation. Am J Pathol 1985;120:112-120. Johnson KJ, Fantone JC, Kaplan J, et al. In 2129 damage to rat lungs by oxygen metabolites J Clin Invest 1981;67:983-993. Shasby DN, Shasby SS, Peach MJ. Granulocytes and phorbol myristate acetate increase permeability to albumin in cultured endothelial monolayers and isolated perfused lungs. Am Rev Resp Dis 1983; 127:72-76. Ayars GH, Altman LC, Rosen H, et al. The injurious effects of neutrophils on pneumocytes in zigzg. Am Rev Resp Dis 1984;130:964-973. Martin WJ. Neutrophils kill pulmonary endothelial cells by a hydrogen peroxide-dependent pathway. An in vitro model of neutrophil-mediated lung injury. Am Rev Respir Dis 1984;130:209-213. Fantone JC, Johnson KJ, Till GO. Acute and progres- sive lung injury secondary to toxic oxygen products from leukocytes. Chest 1983;83:465-485. Brigham KL, Meyrick B. Interactions of granulocytes with the lungs. Circ Res 1984;54:623-635. 175. 176. 177. 178. 179. 180. 181. 182. 183. 184. 185. 186. 163 Ward PA, Sulavik MC, Johnson KJ. Rat neutrophil acti- vation and effects of lipooxygenase and cyclooxygenase inhibitors. Am J Pathol 1984;116:223-233. Barst RJ, Stalcup SA, Steeg CN, et al. Relation of arachidonate metabolites to abnormal control of the pulmonary circulation in a child. Am Rev Resp Dis 1985;131:171-177. Fowler AA, Fisher FJ, et a1. Development of the adult respiratory distress syndrome: Progressive alteration in neutrophil chemotactic and secretory processes. Am J Pathol 1984;116:427-435. Zimmerman GA, Renzetti AD, et a1. Functional and meta- bolic activity of granulocytes from patients with adult respiratory distress syndrome. Am Rev Resp Dis 1983;127:290-300. Gee MH, Perkowski SZ, Tahamount MV, et a1. Arachidon- ate cyclooxygenase metabolites as mediators of com- plement-initiated lung injury. Fed Proc 1985; 44:46-52. Mencia-Huerta JM, Razin E, Corey EJ, et a1. Immuno- logic and ionophore-induced generation of Leukotri ene B from mouse marrow-derived mast cells. J Immunol 1983;130:1885-1890. Weiss SJ, Ward PA. Immune complex-induced generation of oxygen metabolites by human neutrophils. J Immunol 1982;129:309-313. Heflin AC, Brigham KL. Prevention by granulocyte depletion of increased vascular permeability of sheep lung following endotoxemia. J Clin Invest 1981; 68:1253-1260. Quie PG, Mills EI. Bacteriacidal and metabolic func- tion of polymorphonuclear leukocytes. Pediatrics 1979;64:719-721. Miller ME. Phagocyte function in the neonate: select- ed aspects. Pediatrics 1979;64:709-712. Wright DG, Gallin JI. A functional differentiation of human neutrophil granules: Generation of C5a by a specific (secondary) granule product and inactivation of C5a by azurophil (primary) granule products. J Immunol 1977; 119:1068-1076. Kawaguchi H, Koiwai N, Ohtsuka Y, et.al. Phagocytosis of latex particles by leucocytes. 1. Dependence of phagocytosis on the size and surface potential of particles. Biomaterials 1986; 7:61-66. 187. 188. 164 Clark RA. Extracellular effects of the myeloperoxi- dase-hydrogen peroxide-halide system. In: Weissmann G (ed) Adv Inflamm Res 1983; 5:1078-1146. Slocombe RF, Derksen FJ, Robinson NE. Responses of Calves to Intratracheal Challenge with Pasteurella haemolytica and Endotoxin. 1985 American College of Veterinary Pathologists Proceedings [abstract]. p 71. "Illllfllll'fllfi'lfliflfiflr